Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
1269 Signal Requirements for the Generation of CD4+ and CD8+ T-Cell Responses to Human Allogeneic Microvascular Endothelium Ruggero Pardi and Jeffrey R. Bender Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 Microvascular endothelium has been implicated as a major target in the rejection of vascularized allografts. In an attempt to dissect the stepwise generation of the T-cell-mediated immune response to microvascular endothelial cells (ECs), we analyzed the requirements for the two major T-cell subsets, CD4+ and CD8, in the triggering of proliferative and cytotoxic responses to allogeneic ECs in vitro. Results demonstrate that resting ECs are unable to stimulate a functional response by purified T, CD4+, and CD8+ cells in the absence of costimulatory signals. T cells and CD8+ cells develop both proliferative and cytotoxic anti-EC responses by the addition of as little as 2 units/ml interleukin-2, 10 units/ml interleukin-1, or irradiated monocytes autologous to the responder lymphocytes, whereas only autologous monocytes are capable of triggering CD4+ T-cell precursors to proliferate and become anti-EC-specific effector cytotoxic T lymphocytes. CD8' cell-mediated anti-EC cytotoxicity is directed toward allogeneic major histocompatibility complex (MHC) class I determinants on ECs and involves recognition by the CD3/T-cell receptor complex and the CD8 molecule on the effector T cells. CD4+ cells can be driven to proliferate, produce interleukin-2, and become anti-EC-specific cytotoxic T lymphocytes despite the lack of detectable membrane MHC class II determinants on the target cells. Chloroquine inhibition experiments demonstrate that autologous monocytes/macrophages are required by CD4+ T-cell precursors for the processing of EC-derived alloantigens and their subsequent presentation in the context of self-MHC molecules. These results are in agreement with the adoptive transfer experiments in experimental allograft models and suggest that the coordinate engagement of cells of the CD4, CD8, and monocyte/macrophage series is more effective than the individual cell subsets in the generation of a functional response to allogeneic nonlymphoid tissues. (Circulation Research 1991;69:1269-1279) T he continuous exposure of vascular endothelium to circulating lymphocytes, together with its ability to express pleiotropic major histocompatibility complex (MHC) antigens in pathological conditions,1-3 imparts this tissue with the ability to greatly amplify the immune response toward genetically incompatible grafts, or self-antigens, in certain autoimmune diseases.4 Accumulating evidence also supports the notion that microvascular endothelium can be involved as a primary target of functional rejection in vascularized allografts. EndoFrom the Scientific Institute S. Raffaele (R.P.), Milan, Italy, and the Department of Internal Medicine (R.P., J.R.B.), Yale University School of Medicine, New Haven, Conn. Supported by National Institutes of Health grant HL-43331. J.R.B. is the recipient of a National Institutes of Health Clinical Investigator Award (K08 HL-02126/01). Address for correspondence: Dr. J.R. Bender, Division of Cardiology, 3-FMP, P.O. Box 3333, Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06510. Received October 24, 1990; accepted June 21, 1991. thelial cell (EC) injury has been demonstrated in early stages of rejection, with perivascular mononuclear cell infiltrates and interendothelial cell gaps rapidly progressing to blood vessel lining disruption.5 The immunohistological demonstration of EC involvement in allogeneic as well as self-directed immune response has prompted a number of in vitro studies focused on the immunogenicity of vascular endothelium. Our group6 and other groups7 have shown that cultured ECs, either constitutively or after exposure to lymphocyte-derived cytokines, express functionally normal MHC determinants, in addition to polymorphic surface antigens largely restricted to this tissue.8-1' In agreement with these findings, ECs have been shown to elicit proliferative and cytotoxic responses by allogeneic lymphocytes,12 to present soluble antigens to T cells,13"4 and to secrete soluble factors with immunoregulatory activity, such as interleukin (IL)-1,15 IL-6,16 and granulocyte/macrophage colony stimulating factor.'7 1270 Circulation Research Vol 69, No 5 November 1991 In an attempt to investigate the stepwise generation of the T-cell-mediated immune response toward allogeneic microvascular endothelium, we studied the signal requirements for the two major T-cell subpopulations, CD4+ and CD8+, in the induction of a functional response to ECs in vitro. Results presented here demonstrate that resting microvascular ECs are poorly immunogenic to purified T cells in vitro and that a complex interaction between various T-cell subsets and autologous accessory cells (ACs) is required to trigger a T-cell-mediated immune response toward this allogeneic tissue. Furthermore, the data suggest that CD4+ cytotoxic T cells, recognizing processed EC alloantigens, may play a role in the cascade of events leading to EC damage in vivo in vascularized allograft rejection. Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 Materials and Methods Microvascular EC Cultures Stable cultures of human microvascular ECs were established as previously described.18 Briefly, fresh discarded preputial skin from randomly selected anonymous newborns was treated with 0.3% trypsin and 1.0% EDTA for 1 hour at 37°C, after which the microvascular ECs were extruded, plated in microvascular EC culture medium,18 and grown to confluence. On reaching confluence, the cultures were passaged and used for functional assays within the fifth passage. Cells were identified as ECs by standard light microscopy and immunofluorescence techniques as previously described18 and discarded if they contained >1% contaminating cell types. Experimental ECs were considered in resting conditions based on 1) the lack of surface class II HLA expression or monolayer reorganization and 2) background [3H]thymidine incorporation levels. Stable cultures of autologous fibroblasts were established by plating a small aliquot of cells obtained in the EC isolation procedure into a 35-mm dish in the presence of 10% peripartum serum but without EC selection components.18 Monoclonal Antibodies Anti-Leu-2a (clone SK1, immunoglobulin [Ig] G1) binds to the CD8 antigen present on cytotoxic/suppressor T cells; anti-Leu-3a (clone SK.3, IgG1) reacts with the CD4 antigen on helper/inducer T cells; anti-Leu-4 (clone SK.7, IgG1) reacts with CD3 complex present on mature T cells; anti-Leu-2a, -3a, and -4 monoclonal antibodies were purified over Sepharose protein A columns from crude ascites; W6/32, an IgG2a reacting with a framework-specific determinant on HLA-A, -B, and -C molecules was provided by Dr. Peter Parham, Stanford University; CA-141, an IgGI that reacts with a monomorphic determinant on HLADR, was provided by Dr. H.O. McDevitt, Stanford University; L243 is an IgG2a that reacts with a different HLA-DR framework-specific epitope than CA141 and was provided by Dr. R. Levy, Stanford University; anti-lymphocyte function associated (LFA)-1 antigen (CD11a) (clone TS.1.22) and antiLFA-3 antibodies, both IgGl, react with broadly distributed antigens on hematopoietic tissues (LFA-1) and on both hematopoietic and nonhematopoietic tissues (LFA-3) and were provided by Dr. A. Krensky, Stanford University; anti-Leu-M3 (clone MOP9, IgG2b) reacts with monocytes/macrophages and was purchased from Becton Dickinson, Mountain View, Calif. Anti-Leu-10 (clone SK10, IgGl), which reacts with a polymorphic determinant of HLA-DQ expressed in linkage disequilibrium with DR-2, -4, and -6, and anti-HLA-DP (clone B7/21, IgGl), which reacts with a framework-specific determinant of HLADP, were purchased from Becton Dickinson. Reagents Recombinant human interleukin (IL)-la and -1,3 were gifts from S. Gillis, Immunex Corp., Seattle, Wash. Recombinant human interleukin (IL-2) was obtained from Cetus Corp., Emeryville, Calif. Chloroquine was purchased from Sigma Chemical Co., St. Louis, Mo. Flow Cytometry Immunofluorescent staining was performed as described elsewhere in detail.19 All dilutions, washing, and incubations were performed in cold phosphatebuffered saline (0.1 M, pH 7.3) containing 0.1% NaN3. Cytofluorographic analysis was performed by means of an Ortho Cytofluorograph System 50H and a FACStar (Becton Dickinson), using fluorescein isothiocyanate-conjugated or phycoerythrin-conjugated monoclonal antibodies. Isolation of Mononuclear Cell Subpopulations Leukocyte-rich fractions from healthy Red Cross blood donors were randomly selected. Peripheral blood mononuclear cells were isolated by FicollHypaque gradient centrifugation, after which they were rosetted by a single-step rosetting method20 using neuraminidase-treated sheep erythrocytes to obtain erythrocyte-rosette-forming cells. Erythrocyte-rosette-forming cells were further depleted of monocytes and B cells by plastic adherence and passage over a nylon wool column. This yielded a T-cell population that was >95% CD3+ and <0.3% Leu-M3+ when analyzed by flow cytometry. These T-cell-enriched populations were also functionally depleted of monocytes, as demonstrated by the lack of proliferative response to 1 ,ug/ml phytohemoagglutinin (not shown), which effectively induces [3H]thymidine incorporation if the peripheral blood mononuclear cells are not monocyte-depleted. The T cells were used to isolate CD4- (>95% CD8+) and CD8- (>95% CD4+) subsets by negative selection using a panning technique.21 Both subsets contained <0.1% Leu-M3+ cells. To obtain monocyte-enriched ACs, the nonrosetting fraction from the erythrocyte-rosette-forming cell isolations was adhered to plastic flasks (60 minutes at 37°C) and recovered with 1% EDTA (10 Pardi and Bender T-Cell Alloreactivity to Microvascular Endothelium minutes at 4°C) and gentle scraping. The resultant population was >85% Leu-M3+ and had a characteristic monocyte appearance by light microscopy and cytofluorographic light scatter. CD4' Cell Clone Generation Peripheral blood mononuclear cells obtained from a donor were rosetted and monocyte-depleted as described above. CD4+ cells (lx 106) were plated in Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 each well of a 24-well plate (Costar Corp., Cambridge, Mass.) containing confluent monolayers of either the unstimulated MHC class II negative EC30 line or the y-interferon-treated (60 units/ml) MHC class LI positive EC45 line in medium containing 10% pooled human serum, 10% lymphoblastoid cell lineconditioned medium, 20 units/ml IL-2, and 2x i0' irradiated autologous (30 Gy) ACs. The T cells were refed with fresh medium every 3-4 days and added to fresh Ta negative or positive EC monolayers at weekly intervals, at which time they also received fresh, irradiated autologous ACs. After 28 days, the initial populations had gone through a 30-40-fold expansion and were >95% CD4+. Cells were then cloned by limiting dilution into microtiter wells at 0.5 cells/ well and propagated under identical culture conditions. For functional analyses, cloned cells were "rested" in medium without IL-2 or stimulator cells for 18-24 hours before the assay. Proliferation and T-Cell Growth Factor Production Assays Lymphocyte proliferation was evaluated on the basis of [3H]thymidine incorporation. Briefly, 5 x 104 lymphocytes were cultured in microtiter well plates with or without irradiated (30 Gy) allogeneic stimulator ECs (lx 104/well), irradiated autologous ACs (1 x 104/well), and IL-1 or IL-2 at the indicated concentrations. Proliferation was assessed at day 7 by the addition of 1 ,uCi/well [3H]thymidine 18 hours before harvest on glass fiber filters. When testing proliferation of cloned cells, identical cell ratios were used, including allogeneic ACs in one combination tested, but the cultures were carried out for 72 hours and pulsed with [3H]thymidine 12 hours before harvest. All samples were performed in triplicate. T-cell growth factor activity was quantitated using [3H]thymidine incorporation during dose-dependent proliferation of the IL-2-dependent murine T-cell line HT-2, as described elsewhere.22 T-Cell-Mediated Cytotoxicity Assay T, CD4+, or CD8+ cells were cultured in 12-well plates (1 x 106 cells/well) containing confluent EC monolayers (2x i0' cells/well), with or without irradiated autologous ACs, 2 units/ml IL-2, or 10 units/ml IL-1 in 2 ml medium containing 10% pooled human serum. In some experiments, culture supernatants from ECs or 72-hour CD4+/EC cocultures were added at a 1:1 ratio to the indicated coculture media. After 7 days, the suspension cells, which were always >90% viable (both CD4+ and CD8+ cells), 1271 were recovered and used as effectors in 51Cr release assays. These assays were performed in flat-bottomed microtiter wells with confluent EC monolayers, either autologous or allogeneic to the stimulating EC line, which had been labeled with 2 ,Ci/well 5`Cr for 2 hours at 37°C. Spontaneous and maximum release was determined by incubating labeled targets in culture medium or 0.7% Triton X-100 in glycerol/ water, respectively. Mean corrected percent lysis in triplicate culture was calculated as follows: cytotoxicity (%)=(mean experimental cpm-spontaneous release cpm)/(maximum release cpm-spontaneous release cpm)x100. Spontaneous release was always <20% of maximum release. In some experiments, monoclonal antibodies (10 ,ug/ml) were added either directly to the 4-hour assay or used to pretreat either the effector or target cells for 30 minutes at 37°C, after which unbound antibody was washed out and the assay was performed. In other experiments, ACs were treated with 500 ,uM chloroquine at the onset of the 7-day stimulation period. In preliminary experiments, 500 ,uM chloroquine gave the most efficient functional inhibition without affecting cell viability (not shown) and was therefore used in subsequent experiments. ACs were suspended at 1 x 106/ml in RPMI medium containing 500 ,uM chloroquine for 30 minutes at 37°C, after which they were flooded with 10% fetal calf serum, washed twice, irradiated (30 Gy), and added to the responder cells for the 7-day culture. Alternatively, ACs were pulsed overnight with the relevant ECs at a 1:10 (AC: EC) ratio and chloroquine-treated the next day before the initiation of the 7-day culture. In other experiments, the natural killer (NK)-sensitive line K562 was 51Cr-labeled (50 LCi/106 cells) and used as a target in cytotoxicity assays. Results Basal Requirements for the Generation of Anti-EC T-Cell-Mediated Immune Response Primary sensitization in vitro of highly purified T cells or T-cell subsets with allogeneic microvascular ECs is not sufficient to stimulate a detectable anti-EC cytotoxic response (Figure 1). This is not dependent on the particular allogeneic combination used or the time required to generate cytotoxic T lymphocytes (CTLs), as identical results were obtained using multiple genetically unrelated EC lines as stimulators or prolonging the primary lymphocyte/EC coculture for as long as 3 weeks (not shown). These results are at variance with the classical mixed lymphocyte reaction assay, in which mononuclear lymphoid cells are used as stimulators. We reasoned that either unstimulated ECs did not display immunogenic MHC antigens or that they were unable to provide additional signals required for triggering the proliferation and differentiation of allospecific anti-EC precursor T cells. To address these points, primary lymphocyte/EC cocultures were performed in the presence of low concentrations of 1272 Circulation Research Vol 69, No 5 November 1991 60 45 T cells 60 CD8+ cells 60 CD4+ cells control IL-1 10 u/mI 2 u/mI IL-2 autologous AC 45 -45- 010 Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 0 0 0 T cells or T-cell rosetted purified subpopulations cocultured with allogeneic endothelial cells for FIGURE 1. Bar graphs showing 7 days, with or without interleukin (IL)-1, IL-2, or autologous accessory cells (AC), after which the lymphocytes were recovered and used as effectors in a 4-hour cytotoxicity assay, using the relevant endothelial cell line as a target. Similar results were demonstrated in four different experiments all of which were performed at an effector to target ratio of 50:1. Error bars represent standard deviations from the mean of triplicate samples. recombiant lymphokines or irradiated AC-enriched populations autologous to the responder T cells. Our results demonstrate that these additional signals are indeed capable of driving a functional anti-EC cytotoxic response by T cells (Figure 1). Moreover, they indicate that the minimal requirements for an anti-EC response by CD8+ and CD4+ T-cell subsets are distinct, since the former subset can be induced to become CTLs by the addition to the cultures of 2 units/ml IL-2, 10 units/mi IL-1, or, to a lesser extent, autologous ACs (Figure 1), whereas only the presence in the primary culture of irradiated autologous accessory cells can stimulate an anti-EC cytotoxic response by purified CD4+ T lymphocytes (Figure 1). As previously shown,23 IL-2 concentrations >5 units/ ml, although more efficient in driving anti-EC CD8+ T-cell-mediated cytotoxicity, would increase the amount of nonspecific killing, as detected by using the K562 target cell line (not shown). In contrast, IL-2 could not generate an anti-EC CD4+ CTL response even at concentrations as high as 50 units/ml (not shown). EC culture or CD4+/EC coculture supernatants had no effect on the ability to derive EC-specific CD4+ cytotoxic T cells in the presence of autologous ACs or EC-specific CD8+ cytotoxic T cells in the presence of IL-2, ruling out that the inability of CD4+ T cells to respond to ECs in the absence of ACs was due to the elaboration of relevant suppression of inhibitory factors in the experimental conditions (not shown). When the T-cell or T-cell subset proliferation was evaluated under the same experimental conditions, similar results were observed (Table 1). IL-2 (2 units/ml) was the most efficient costimulatory signal for CD8+ lymphocytes, whereas autologous ACs were essential for driving an anti-EC proliferative response by purified CD4+ cells (Table 1). Specificity of the Anti-EC Cytotoxic Response by CD4+ and CD8+ T Lymphocytes To further investigate whether the costimulatory signals required to generate anti-EC cytotoxic T cells were nonspecifically triggering lytic mechanisms by the effector cells and whether EC alloantigens were actually involved in the generation of the response, multiple genetically unrelated EC lines were used as targets in the final cytotoxicity assay, in addition to the NKsensitive line, K562. The results (Figures 2a and 2b) demonstrate that the two most efficient pathways for generating an anti-EC cytotoxic response by T-cell subsets (i.e., CD8+ cells plus 2 units/ml IL-2 and CD4+ cells in the presence of irradiated autologous ACs) display a high degree of target-related specificity. This is further demonstrated by crossover killing experiments, where CD8+ (Figure 2c) or CD4+ (Figure 2d) cells from another donor were cocultured with a different EC line (EC44) than that shown in Figures 2a and 2b. The absence of reciprocal killing (CD8 or CD4 anti-EC45 lyses EC45 better than EC44, and CD8 or CD4 anti-EC44 lyses EC44 better than EC45) indicates that the stimulating EC alloantigens are, in fact, responsible for target specificity and that differential target sensitivity is not a major component of the noted phenomenon. Furthermore, fibroblasts syngeneic to the target EC44 line are also lysed by the CD8+ and CD4+ effector cells (Figures 2c and 2d), suggesting that the target antigens are shared by ECs and fibroblasts. With fibroblast targets, the percent cytotoxicity is lower at each effector to target cell (EiT) ratio. A greater or more homogeneous expression of the putative alloge- Pardi and Bender T-Cell Alloreactivity to Microvascular Endothelium 1273 TABLE 1. Allogeneic Endothelial Cell-Induced Proliferative Response and Interleukin-2 Production by T Cells and T-Cell Subsets Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 Proliferation (cpm) Responders HT-2 cell responset (cpm) Cofactors* 310±40 None T cells 399+78 T cells 5,310±617 IL-2 9,911 +799 T cells 6,601±+788 IL-1 9,002+786 T cells Auto ACs 8,356+1,005 19,566±2,100 None 206±46 CD4+ cells 602±90 IL-2 701±109 2,654±769 CD4+ cells 987± 110 1,292±264 IL-1 CD4+ cells 8,166±1,203 20,001±2,504 Auto ACs CD4+ cells 504±70 320±48 None CD8+ cells IL-2 3,899±526 CD8+ cells 8,543+1,320 2,902±456 IL-1 CD8+ cells 3,700+601 ND ND Auto ACs CD8+ cells Values are mean±SEM of triplicate cultures. HT-2, interleukin (IL)-2-dependent murine T-cell line; Auto ACs, autologous accessory cells; ND, not done. Purified T cells or CD4+ and CD8+ T cells (1 x 105) were cocultured with irradiated allogeneic endothelial cells for 6 days, after which 1 ,Ci [3H]thymidine was added, and cultures were harvested 18 hours later. *Lymphokine concentrations were 2 units/ml for IL-2 and 10 units/ml for IL-1. Monocyte-enriched populations autologous to the responder lymphocytes (Auto ACs) were irradiated and added to the cultures at a monocyte to T-cell ratio of 0.2: 1. T-cell response to Auto ACs, as measured by [3H]thymidine incorporation, never exceeded 2,000 cpm. tT-cell growth factor activity was assessed with the HT-2 cell bioassay. HT-2 cells (2 x 104) were cultured for 24 hours in the presence of 100 ,ul of 72-hour culture supernatant from the indicated cell combinations. Samples were labeled with 1 ,uCi [3H]thymidine 4 hours before harvesting the cells. neic target molecule(s) may be present on ECs, although it is possible that fibroblasts are relatively more resistant to T-cell-mediated lysis. Cell Surface Antigens Involved in T-Cell-Mediated Anti-EC Cytolysis To identify the surface molecules involved in the anti-EC cytotoxic response by T-cell subsets, soluble monoclonal antibodies directed toward a number of lymphocyte or EC membrane antigens were added to the final cytotoxic assay. Figure 3a shows that antibodies recognizing the CD3-T cell receptor complex, the CD8 antigen, and MHC class I framework determinants were effective, to a variable extent, in blocking the CD8+ cell-mediated anti-EC cytotoxic activity. Experiments performed by pretreating either the effector or the target cells with the indicated antibodies demonstrated that MHC class I determinants on the surface of ECs, but not of lymphocytes, were involved in the recognition process leading to EC lysis by CD8+ T cells (not shown). When the same panel of antibodies was added to CD4+ cells that had been cultured with ECs, only antibodies against the CD3iTcR complex or the CD4 molecule could effectively inhibit the effector function of this cell subset. Anti-class I, anti-HLA-DP, antiHLA-DQ, and two anti-HLA DR antibodies recognizing distinct epitopes of the heterodimer did not block this activity at E/T ratios as low as 6: 1 (Figure 3b, data not shown). ECs used as stimulators in the generation of EC-specific T cells or T-cell subsets were, in fact, always MHC class II antigen positive at the end of the 7-day coculture period (data not shown). However, to further confirm the lack of involvement of EC Ia antigens in the CD4+ T-cell-mediated anti-EC response, resting ECs, in conditions identical to the assay targets, were stained with antibodies recognizing monomorphic HLA-DP, -DQ, and -DR epitopes. Figure 4 shows that, as previously demonstrated with large vessel-derived ECs,1 microvascular ECs used as targets in the cytotoxicity assay do not express detectable levels of membrane MHC class II antigens, whereas surface MHC class I determinants are constitutively present at a high density. When induced, EC membrane HLA class II is detectable on the cell surface (Figure 4d). Role ofAutologous ACs in the Generation of CD4+ T-Cell-Mediated Anti-EC Response The above results demonstrate an essential role for irradiated autologous ACs in the generation of an anti-EC response by CD4+ lymphocytes, but their precise role is uncertain. Conceivably, these cells produce soluble factors, distinct from IL-1, that act as costimulatory signals in the expansion and differentiation of CD4+ allospecific T-cell precursors. Alternatively, ACs might process and present EC-derived membrane alloantigens in an immunogenic manner to autologous CD4+ lymphocytes. To explore these possibilities, ACs were pulsed overnight with the stimulator EC line and then treated with the weak base chloroquine, an agent known to block exogenous antigen processing by inhibiting the catabolism of peptides in intracellular acidic vesicles.24,25 Chloroquine-treated alloantigen-pulsed ACs were then used as costimulators in the CD4+ lymphocyte/EC primary culture. The results (Figure 5) demonstrate that treatment of autologous ACs with chloroquine (500 gM) before the addition of ECs abrogated the subsequent generation of anti-EC cyto- Circulation Research Vol 69, No 5 November 1991 1274 a EC 45 O EC44 0 EC 33 K562 b 451 a 4' 0) .0 0 15 0 15 50:1 25:1 12.5:1 6:1 Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 25:1 12.5:1 E/T ratio EIT ratio 60 60 -d FIGURE 2. Graphs showing microvascular endothelial cell (EC) line 45 cocultured for 7 days with CD8+ T cells in the presence of 2 units/ml interleukin-2 (panel a) or CD4+ T cells in the presence of irradiated autologous accessory cells (accessory cell to lymphocyte ratio, 0. 2:1) 6:.1 (panel b). EIT ratio, effector to target cell ratio. After 1 week, the lymphocytes were recovered and used as effectors in a 4-hour cytotoxicity assay, with the relevant or irEC 44 relevant EC lines as targets, as well as the Cl EC45 o 5 natural killer target K562. The identical EC o cocultures and cytotoxicity experiments * K.562 were performed with CD8' T cells (panel * Fib 44 c) or CD4+ T cells (panel d) obtained from a different donor than for panels a and b and stimulated with EC line 44. Fib 44 is syngeneic to EC44. EC62 45 45 C) Co .0 0 ,, 30 30 KR 01 --a 1 15 0 50:1 25:1 12.5:1 6:1 E/T ratio o N ,v -a 11 - - 50:1 25:1 12.5:1 E/T ratio toxic CD4+ T cells, whereas the drug was ineffective when used after the ACs had been pulsed overnight with allogeneic ECs. This strongly suggests that an intracellular alloantigen processing step, carried out by ACs, is a prerequisite for the sensitization of anti-EC allospecific CD4+ precursor T lymphocytes. Analysis of the CD4' T-Cell-Mediated Anti-EC Response at the Clonal Level The restriction elements responsible for the induction of anti-EC-specific, CD4+ CTLs were further analyzed by generating CD4+ cytotoxic T-cell clones, under limiting dilution conditions, using MHC class II negative allogeneic EC lines as stimulators in the presence of ACs autologous to the responder lymphocytes. Data obtained with one representative clone, designated AT.30.1, are illustrated in Table 2. This clone is a CD3+, CD4+, and a1if3 T-cell clone, as demonstrated by its reactivity with the monoclonal 6:1 antibody WT31, which recognizes an epitope on the antigen receptor complex of a1lp T cells. Also, the AT.30.1 clone does not kill K562, nor does it express classical natural killer cell markers; thus, it is not a natural killer or natural killer-like T-cell clone (not shown). The data demonstrate that the same cloned cell line is capable of EC-specific, self-MHC-restricted proliferation in the presence of autologous monocytes as a source of ACs and of cytotoxicity to the relevant EC line. IL-2 also is extremely efficient in triggering a proliferative response by cloned CD4' cells. Of note, ECs alone were extremely inefficient in driving a proliferative response by the AT.30.1 clone, although they could serve as a sensitive target in the cytotoxic assay. Three additional clones (out of 36 anti-EC CD4+ clones tested) had functional properties similar to the AT.30.1 clone; the remainder were not capable of anti-EC-specific cytotoxic activity (not shown). Anti- Pardi and Bender T-Cell Alloreactivity to Microvascular Endothelium 0 % specific lysis 25 50 % specific lysis irrel.Gi irrel.G23 anti-CD3 anti-HLA-I anti-HLA-DR anti-CD8 anti-C D4 a b Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 FIGURE 3. Bar graphs showing microvascular endothelial cells cocultured for 7 days with CD8+ T cells in the presence of 2 units/ml interleukin-2 (panel a) or CD4+ T cells in the presence of irradiated autologous accessory cells (accessory cell to lymphocyte ratio, 0.2: 1) (panel b). irrel. G, and G2, irrelevant immunoglobulin GJ and G2a. After 1 week; the lymphocytes were recovered and used as effectors in a 4-hour cytotoxicity assay with the relevant endothelial cell target, in the presence of saturating concentrations (10-25 pglml) of the indicated monoclonal antibodies. Results are representative of four separate experiments, allperformed at effector to target cell ratios of 25:1. 1275 body inhibition studies of this clone confirmed the results obtained at the bulk culture level, demonstrating the involvement of the CD3/T-cell receptor complex and the CD4 molecule in both the inductive and effector phases of the anti-EC response by CD4' lymphocytes (Table 2). Anti-MHC class I and two anti-HLA-DR antibodies recognizing nonoverlapping epitopes of the DR molecule were again unable to inhibit the allogeneic EC-specific cytotoxicity mediated by cells of the CD4' sublineage. When y-interferon-treated, MHC class II-expressing ECs were used as stimulators to generate allospecific CD4+ CTL clones, the recognition pattern of the effector cells was at variance with the results observed using MHC class II negative stimulators. Under these circumstances, the majority of the CTL clones tested appeared to recognize MHC class II determinants, as shown by the absence of killing of Ia negative EC targets and by the inhibition of class IL+ EC target lysis observed using anti-class II monoclonal antibodies (Table 3). Discussion Donor microvascular endothelium is likely to be involved both as a stimulus and as a target in FIGURE 4. Cell-sorter immunofluorescent analysis of microvascular endothelial cells cultured in standard medium'8 (panels A-C and E) or medium containing 100 units/ml yinterferon (panel D), comparing irrelevant immunoglobulin GJ staining with specific endothelial cell surface major histocompatibility complex expression. Monoclonal antibodies used were fluorescein isothiocyante (FITC)-CA-141 (anti-HLADR, panel A), FITC-Leu-10 (antiHLA-DQ, panel B), FITC-anti-DP (anti-HLA-DP, panel C), FITCCA-141 (panel D), and FITC-W6/32 (anti-HLA class I, panel E). Ten thousand cells were analyzedfor each sample. 1276 Circulation Research Vol 69, No 5 November 1991 Exp. 1 60 45 45 k, Exp. 2 60 r- 2 30 15 15 [- t 0 50:1 25:1 12.5:1 EIT ratio 1 6:1 0 50:1 25:1 12.5:1 6:1 EIT ratio Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 FIGURE 5. Graphs showing punified CD4+ T cells cocultured for 7 days with allogeneic endothelial cells and autologous accessory cells that had been pretreated with medium (D), 500 pM chloroquine (o), or 500 MM chloroquine after ovemight pulsing with the relevant endothelial cell line (o). EIT ratio, effector to target cell ratio. After 1 week, the lymphocytes were recovered and used as effectors in a 4-hour cytotoxicity assay with the relevant endothelial cell target. vascularized allograft rejection, since it forms a contiguous barrier, bearing surface donor alloantigens, to circulating recipient lymphocytes. Immunohistological data in human vascularized grafts undergoing rejection support this notion, because they frequently demonstrate widespread microvascular damage, which is initially seen in those venules, arterioles, and small veins enveloped by lymphocytes.26 However, analysis of mononuclear cell infiltrates in rejected allografts is not sufficient to determine which lymphocyte subset is primarily responsible for damaging the donor tissue, nor does it clarify the sequence of events underlying the rejection process. For this reason, we analyzed the signals and restriction elements used by unprimed lymphocyte subsets to generate a functional response in vitro to allogeneic ECs of microvascular origin. The results demonstrated that resting microvascular ECs are unable to trigger a functional response by allogeneic monocyte-depleted T cells. These observations confirm the data obtained using ECs derived from large veins as stimulators27 and suggest that T cells, unlike their functional response to genetically unrelated lymphoid cells, require additional signals to proliferate and differentiate into allospecific effectors. Because ECs express surface class I MHC determinants with a density comparable to lymphoid cells,' it is likely that their inability to stimulate allogeneic T cells reflects either defective "accessory" cell function (i.e., soluble factor production or antigen processing capabilities) or insufficient expression of MHC class II allodeterminants to be recognized by allospecific T cells. The induction of alloantigen-specific T-cell proliferation and cytotoxicity to allogeneic ECs in the presence of low concentrations of IL-1 or IL-2 (Figure 1 and Table 1) would suggest that surface alloantigens displayed by ECs are normally recognized by precursor T cells, whereas the IL-1 mediated activation pathway involved in the afferent limb of the T-celldependent immune response28 is defective. This conclusion is supported by the experiments performed with CD8' cells (Figure 1 and Table 1), which were induced to lyse the sensitizing EC line if as little as 2 units/ml IL-2 were present in the 7-day cultures. This result suggests that CD8+ cells express TABLE 2. Proliferative and Cytotoxic Activity of Allogeneic Endothelial Cell-Specific CD4+ T-Cell Clone (AT.30.1) Proliferationt Inhibition (%) cpm ... 2,812 ... 66,332 ... 5,835 6,002 0 25,331 2.5 24,698 38.2 15,639 44.0 14,208 0 26,112 33.6 16,838 45.2 13,901 Cytolysist Inhibition (%) Lysis (%) Stimulators Cofactors Blocking MAb* + ... ... ... None + ... ... ... IL-2 (2 units/ml) ... Auto ACs + ... Allo ACs 0 ... + 78.0 Auto ACs 71.8 8.0 + Irrelevant Auto ACs 53.8 39.0 + Anti-CD3 Auto ACs 28.4 63.6 + Anti-CD4 Auto ACs + 64.0 18.0 Auto ACs W6/32 + CA-141 74.9 4.0 Auto ACs + L243 71.0 9.0 Auto ACs ND ... 74.0 + Anti-LFA-1 20.3 Auto ACs 71.4 + Anti-LFA-3 ND ... 8.5 Auto ACs MAb, monoclonal antibody; IL-2, interleukin-2; Auto ACs, autologous accessory cells; Allo ACs, allogeneic accessory cells; LFA, lymphocyte function associated; ND, not done. *The indicated antibodies (10 ,ug/ml) were added at the beginning of the culture or during the 4-hour cytotoxic assay. tAT.30.1 cloned cells were recovered, rested for 18 hours in the absence of stimulating antigen or IL-2, and cultured (5 x 104) for 72 hours in the presence of the indicated cofactors, with or without irradiated endothelial cells (lx 104) from the line originally used as stimulator. Irradiated autologous (auto AC) or allogeneic (allo AC) monocytes were added at a monocyte to T-cell ratio of 0.2: 1. Twelve hours before harvesting the cultures, 1 gCi [3H]thymidine was added. *Cloned AT.30.1 cells were washed and added to 5`Cr-labeled endothelial cells from the original stimulator line at a T-cell to endothelial cell ratio of 50: 1, with or without 10 ,ug/ml of the indicated antibodies. Four hours later, `tCr release was evaluated in the supernatants, and specific lysis was calculated as described in "Materials and Methods." Pardi and Bender T-Cell Alloreactivity to Microvascular Endothelium 1277 TABLE 3. Cytotoxic Activity of CD4+ T-Cell Clones Induced by Major Histocompatibility Complex Class II Allogeneic Endothelial Cells Ia+ EC45t Ia- EC45 Clone Blocking MAb* Lysis (%) Inhibition (%) Lysis (%) Inhibition (%) AT.45.3 ... 65.0 0 8.1 0 CA-141 28.1 56.8 7.5 9.7 W632 63.8 1.9 8.3 0 ... AT.45.8 58.3 0 2.1 0 CA-141 19.3 66.9 3.0 0 W632 60.0 0 2.6 0 MAb, monoclonal antibody; EC, endothelial cell. EC45-specific cloned cells were recovered, rested for 24 hours in the absence of stimulating cells, and assayed for cytotoxic activity against 51Cr-labeled EC45 cells at a T-cell to EC ratio of 25: 1. *The indicated MAbs (10 ,tg/ml) were added during the 4-hour chromium release assay. tTo induce major histocompatibility complex class II expression, EC45 cells were pretreated for 72 hours with 60 units/ml y-interferon before the cytotoxicity assay. Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 functional IL-2 receptors on recognition of ECspecific alloantigens but that they do not progress into the differentiation pathway leading to allospecific effector CTLs. The presence of IL-1 or autologous ACs could, therefore, provide the additional signals required for the production of IL-2 by CD8+ cells,29 thus driving an autocrine allospecific response by this lymphocyte subset. The demonstration of an autocrine loop, driven by allogeneic class I MHC antigens and entirely mediated by CD8+ lymphocytes, has precedent, as "helper-independent" cytotoxic lymphocytes have been described both in murine and human experimental systems.23,30 Blocking experiments (Figure 3a) carried out in the final lytic assay using allogeneic EC-primed CD8+ cells as effectors demonstrated that, as expected, membrane class I MHC allodeterminants on ECs represent the major target and presumed stimulus for this lymphocyte subset. The CD3/TcR and CD8 molecular complexes on the effector cells appear to be involved in the recognition and subsequent lysis of the stimulating EC line. These findings confirm observations made by other investigators'4 and indicate that class I MHC antigens displayed by ECs are fully capable of triggering allospecific CD8+ CTL precursors. This functional correlation between CD8+ cells and EC class I MHC alloantigens is consistent with the molecular constraints originally described in the classical mixed lymphocyte reaction model3' and subsequently confirmed by functional studies performed on lymphocytes infiltrating class I MHC incompatible grafts.32-35 The analysis of the CD4+ cell-mediated anti-EC response provided interesting and somewhat unexpected results. The existence of CD4+ cells capable of differentiating into cytotoxic effectors has been clearly demonstrated when allogeneic MHC class II antigen positive cells have been used as stimulators,36 and it was confirmed in the present study using y-interferon-treated Ia-positive ECs as stimulators (Table 3). Nonetheless, the likelihood of detecting such activity in CD4+ lymphocytes cultured for 1 week with allogeneic ECs seemed low, since HLA- DP, -DQ, and -DR antigens are not detectable on microvascular ECs under these conditions (Figure 4). Clonal analysis of the CD4+ cell-mediated anti-EC cytotoxic response demonstrates that, at least in some instances, the same cell can be induced to proliferate in response to the relevant EC alloantigens and to differentiate toward the final effector CTLs (Table 2). The presence in the sensitizing culture of autologous ACs appears to be a condition necessary and sufficient for generating a detectable response by CD4+ cells in our experimental model. The activity of autologous ACs in the culture system is not replaceable by IL-1 or IL-2 (Figure 1 and Table 1), but it is completely abrogated by the treatment of ACs with 500 ,uM chloroquine before alloantigen pulsing (Figure 5). This strongly suggests that intracellular processing of allogeneic EC-derived antigens is required to trigger the recognition of allogeneic EC determinants by CD4+ lymphocytes. Results obtained with an allogeneic EC-specific CD4+ clone (Table 2) suggest that self-MHC class Il-expressing autologous ACs, in addition to processing EC-derived antigens, provide the necessary restriction elements for CD4+ cells, as shown by the inhibition observed with anti-Ia antibodies and by the inability of ACs bearing MHC antigens unrelated to those of the responding T cells to stimulate a proliferative response to the sensitizing EC line. The most likely explanation for these findings is that alloantigen processing and presentation in the context of selfMHC membrane molecules by ACs is a requirement for the stimulation of allospecific responses by CD4+ lymphocytes. This interpretation is further supported by other studies using cell-free minor histocompatibility antigens as stimulators of a T-cell-mediated cytotoxic response in the murine system37-39 and, more recently, by the elegant demonstration that peptide fragments derived from the hypervariable domain of MHC molecules physically associate with membrane Ia antigens on antigen-presenting cells and effectively stimulate an allospecific T-cell response.40'41 The EC-specific alloantigens responsible for the induction of CD4+ cell-mediated anti-EC 1278 Circulation Research Vol 69, No 5 November 1991 responses remain elusive, as resting ECs do not express detectable surface HLA-DP, -DQ, or -DR antigens, which are typically involved in the trigger- Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 ing of CD4+ lymphocyte-mediated activities (Figure 4 and Reference 33). Particularly difficult to understand is how antigen presentation by autologous ACs leads to recognition and lysis of allogeneic ECs. Since no monocytes were added to the cytotoxicity assays and since the resting EC targets lack detectable class II MHC determinants, the possibility exists that cytolysis by the CD4+ CTLs is not specific for classical MHC class II (or class I) molecules. This is surprising in light of the generally accepted view that all or nearly all CD4+ T-cell functions are MHC class LI restricted.42 On the other hand, since we have not yet tested the effects on cytotoxicity of a large panel of MHC class ll-specific antibodies, the possibility remains that ECs may express quantities of MHC class II antigens that are sufficient to trigger cytotoxicity but insufficient to generate an immunofluorescent signal or to stimulate clonal expansion. Alternatively, class II MHC determinants shed either from the activated CD4+ effectors or autologous monocytes may have passively adsorbed onto allogeneic EC targets during the 4-hour chromium release assay and combined with EC alloantigens in a manner recognized by CD4' CTL. There is precedent for passive adsorption of Ia molecules onto T cells that have been incubated with allogeneic non-T mononuclear cells.43 Finally, it is possible that non-MHCencoded EC antigens might play a major role in the sensitization of an allospecific CD4+ CTL precursor. This concept is supported by the existence of ECspecific alloantigens apparently responsible for allograft rejection in HLA-identical individuals44 and by recent reports demonstrating a role for CD4+ cytolytic4546 and noncytolytic47 cells specific for nonMHC-encoded determinants. All of these possibilities remain to be explored. In conclusion, our results indicate that the generation of a functional immune response to allogeneic vascular endothelial cells, which are defective in AC activity, requires a complex interaction between different cell types having specialized functions. In addition to the role of CD4' and CD8' precursor lymphocytes capable of clonal expansion, cytokine production, and allo-EC-specific cytotoxicity, our data underscore the importance of ACs of host origin in the inductive phase of the response to allogeneic tissues. ACs autologous to the responder lymphocytes appear to be critical not only as a source of soluble factors, such as IL-1, involved in the clonal expansion and subsequent differentiation of allo-EC specific precursors but also as elements capable of uptaking and processing the alloantigens required to sensitize CD4+ precursors in association with the appropriate restriction elements.48 These data lend support to the observations made in vivo in experimental allograft rejection models, showing that adoptive transfer of whole mononuclear cell populations to a sublethally irradiated host is consistently more effective in promoting allograft rejection than selected lymphocyte subsets of the helper or cytotoxic lineage (for review, see Reference 26). Moreover, they provide evidence for the existence of a sequential activation process, involving ACs and CD4+ and CD8' T cells, in the response to nonlymphoid allogeneic tissues, as suggested by histological data that shows the serial appearance of activated cells belonging to these lineages in cellular infiltrates associated with graft rejection.49,50 Acknowledgments We gratefully acknowledge Edgar G. Engleman (Stanford University), in whose laboratory some of the experiments were performed, for his advice and critical review of the manuscript. We also thank Jennie Annatone, Georgi Danley, and Kristie Vollono for preparation of the manuscript, Leslie Tackett for assistance with endothelial cell culture, and Silvia Heltai for her help with cytofluorography. References 1. Stastny P, Nunez G: The alloantigens of endothelial cells, in Jaffe EA (ed): Biology of Endothelial Cells. Boston, Martinus Nijhoff Publishers, 1984, pp 377-384 2. Marlene LR, Coles MI, Griffin RJ, Pmerance A, Yacoub MH: Expression of class I and II major histocompatibility antigens in normal and transplanted human heart. Transplantation 1986;41:776-781 3. Hall BM, Bishop GA, Duggin GG, Horvath JS, Philips J, Tiller DJ: Increased expression of HLA-DR antigens on renal tubular cells in renal transplants: Relevance to the rejection response. Lancet 1984;2:247-249 4. DeWaal RMW, Bogman JJ, Maars CN, Cornelissen LMH, Tax WJM, Koene RAP: Variable expression of Ia antigens on the vascular endothelium of mouse skin allografts. Nature 1983;303:426-429 5. Dvorak HF, Mihm MC, Dvorak AM, Barnes BA, Manseau EJ, Galli SJ: Rejection of first set skin allograft in man: The microvasculature is the critical target of the immune response. J Exp Med 1979;105:322-335 6. Pardi R, Bender JR, Engleman EG: Lymphocyte subsets differentially induce class II human leukocyte antigens on allogeneic microvascular endothelial cells. J Immunol 1987; 139:2585-2595 7. Pober JS, Gimbrone MA Jr, Cotran RS, Reiss CS, Burakoff SJ, Fiers W, Ault KA: Ia expression by vascular endothelium is inducible by activated T cells and by human interferon. JExp Med 1983;157:1339-1351 8. Brasile L, Zerbe T, Rabin B, Clarke J, Abrams A, Cerilli J: Identification of the antibody to vascular endothelial cells in patients undergoing cardiac transplantation. Transplantation 1985;40:672-678 9. Blankert JJ, Van ES, Kunz HW, Cramer DV, Paul LC: Genetics of the antibody response to an endothelial transplantation antigen in the rat. Hum Immunol 1986;15:125-129 10. Blankert JJ, Muizert Y, Van ES, Paul LC: Immunogenicity of the non-MHC endothelial antigen EAG-1 in various tissues of the rat. Transplantation 1987;43:736-742 11. Schook LB, Wood N, Mohanakumar T: Identification of human vascular endothelial cell/monocyte antigenic system using monoclonal antibodies. Transplantation 1987;44:412-416 12. Clayberger C, Uyehara T, Hardy B, Eaton J, Karasek M, Krensky AM: Target specificity and cell surface structures involved in the human cytolytic T lymphocyte response to endothelial cells. J Immunol 1985;135:12-20 13. Geppert TD, Lipsky PE: Dissection of defective antigen presentation by interferon-y treated fibroblasts. J Immunol 1987;138:385-392 Pardi and Bender T-Cell Alloreactivity to Microvascular Endothelium Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 14. Kayroska A, Lipsky PE: Dissection of the functions of antigenpresenting cells in the induction of T cell activation. JImmunol 1985;135:2953-2958 15. Moissec P, Cavender D, Ziff M: Production of IL-1 by human endothelial cells. J Immunol 1986;136:2486-2494 16. Leeuwenberg JFM, Von Asmuth EJU, Jeunhomme TMAA, Buurman WA: IFN-y regulates the expression of the adhesion molecule ELAM-1 and IL-6 production by human endothelial cells in vitro. J Immunol 1990;145:2110-2114 17. Sieff CA, Tsai S, Fawer DV: Interleukin-1 induces cultured human endothelial cell production of granulocyte-macrophage CSF. J Clin Invest 1987;79:48-53 18. Bender JR, Pardi R, Karasek MA, Engleman EG: Phenotypic and functional characterization of lymphocytes that bind human microvascular endothelial cells in vitro. J Clin Invest 1987;79:1679-1688 19. Lifson J, Raubitschek A, Benike C, Koths K, Ammann A, Sondel P, Engleman EG: Purified interleukin 2 induces proliferation of fresh human lymphocytes in the absence of exogenous stimuli. J Biol Response Mod 1986;5:61-66 20. Weiner MS, Bianco C, Nussenzweig V: Enhanced binding of neuraminidase-treated sheep erythrocytes to human T lymphocytes. Blood 1973;42:939-944 21. Wysocki LJ, Sato VL: Panning for lymphocytes: A method of cell separation. Proc Natl Acad Sci U S A 1978;75:2844-2849 22. Watson J: Continuous proliferation of murine antigen specific helper T lymphocytes in culture. J Exp Med 1979; 150: 1510-1519 23. Fishwild DM, Benike CJ, Engleman EG: Activation of HLArestricted EBV-specific cytotoxic T cells does not require CD4+(helper) T cells or exogenous cytokines. J Immunol 1988;140:199-204 24. Ziegler K, Unanue E: Identification of a macrophage antigenprocessing event required for I-region-restricted antigen presentation to T lymphocytes. J Immunol 1981;127:1869-1880 25. Ziegler HK, Unanue E: Decrease in macrophage antigen catabolism caused by ammonia and chloroquine is associated with inhibition of antigen presentation to T cells. Proc Natl Acad Sci USA 1981;79:175-179 26. Mason DW, Moris PJ: Effector mechanisms in allograft rejection. Annu Rev Immunol 1986;4:119-153 27. Geppert TD, Lipsky PE: Accessory cell-T interactions involved in anti-CD3 induced T4 and T8 cell proliferation: Analysis with monoclonal antibodies. J Immunol 1986;137: 3065-3073 28. Kaye J, Gillis S, Mizel SB, Shevach EM, Malek TR, Dinarello CA, Lachman LB, Janeway CA: Growth of a cloned helper T cell line induced by a monoclonal antibody specific for the antigen receptor: Interleukin 1 is required for the expression of receptors for interleukin 2. J Immunol 1984;133:1339-1350 29. Mizuochi T, McKean DJ, Singer A: IL-1 as a cofactor for lymphokine-secreting CD8+ murine T cells. J Immunol 1988; 141:1571-1577 30. Isakov N, Biel LW, Bach FH: Induction of class I antigen: Disparate pancreatic islet cell graft rejection by in vivo administration of cloned helper-independent cytolytic T lymphocytes. Transplant Proc 1985;17:727-731 31. Engleman EG, Benike CJ, Glickman E, Evans RL: Antibodies to membrane structures that distinguish suppressor/cytotoxic and helper T lymphocyte subpopulations block the mixed lymphocyte reaction in man. J Exp Med 1981;153:193-202 32. Moreau JF, Vie H, Peyrat MA, Soulillou JP: Function and cell surface markers of cloned T lymphocytes isolated from human renal allograft biopsies. Transplant Proc 1985;17:810-814 33. Flomenberg N, Duffy E, Dupont B: HLA class II specific T lymphocyte clones with dual alloreactive functions. Scand J Immunol 1984;19:237-245 1279 34. Stepkowski SM, Duncan WR: The role of TDTH and T, populations in organ graft rejection: Functional analysis of graft infiltrating T cells. Transplantation 1986;42:406-410 35. Zeei A, Pung J, Zerbe TR, Kaufman C, Rabin B, Griffith BP, Hardesty RC, Duquesnoy RJ: Allospecificity of activated T cells grown from endomyocardial biopsies from heart transplant patients. Transplantation 1986;41:620-625 36. Mayer TG, Lazarovits AI, Boyle LA, Sullivan ER, Leary CP, Fuller AA, Fuller TC, Bhan AK, Colvin RB, Collins AB, Kurnick JT: Functional allospecific T lymphocytes isolated from human renal allograft biopsies. Transplant Proc 1985;17: 816-819 37. Pilarski LM: In vitro cytotoxic T cell responses to cell-free minor histocompatibility antigens: Requirement for antigen presenting cells. Transplantation 1987;43:286-290 38. Golding H, Singer A: Role of accessory cell processing and presentation of shed H-2 alloantigens in allospecific cytotoxic T lymphocyte responses. J Immunol 1984;133:597-660 39. Weinberger 0, Germain RN, Springer TA, Burakoff SJ: Role of syngeneic Ia+ AC in the generation of allospecific CTL responses. J Immunol 1982;129:694-703 40. Parham P, Clayberger C, Zorn SL, Ludwig DS, Schoolnik GK, Krensky AM: Inhibition of alloreactive cytotoxic T lymphocytes by peptides from the 2 domain of HLA-A2. Nature 1987;325:625-627 41. Clayberger C, Parham P, Rothbard J, Ludwig DS, Schoolnik GK, Krensky AM: HLA-A2 peptides can regulate cytolysis by human allogeneic T lymphocytes. Nature 1987;330:763-760 42. Littman DR: The structure of the CD4 and CD8 genes. Annu Rev Immunol 1987;5:561-592 43. Yu DTJ, McCune JM, Fu SM, Winchester RJ, Kunkel HG: Two types of Ia-positive T cells: Synthesis and exchange of Ia antigens. J Exp Med 1980;152:895-908 44. Brasile L, Clarke J, Galouzis T, Cerilli J: The clinical significance of the vascular endothelial cell antigen system: Evidence for genetic linkage between the endothelial cell antigen system and the major histocompatibility complex. Transplant Proc 1985;17:741-743 45. Erb P, Grogg D, Troxler M, Kennedy M, Fluri M: CD4+ T cell-mediated killing of MHC class II positive antigen presenting cells: Characterization of target cell recognition by in vivo and in vitro activated CD4+ killer T cells. J Immunol 1990; 144:790-795 46. Korngold R, Sprent J: Variable capacity of L3T4+ T cells to cause lethal graft vs. host disease across minor histocompatibility barriers in mice. J Eip Med 1987;165:155-166 47. Miconnet I, Huchet R, Bonardelle D, Motta R, Canon C, Garay-Rojas E, Kress M, Reynes M, Halle-Pannenko 0, Bruley-Rosset M: Graft vs. host mortality induced by noncytolytic CD4+ T cell clones specific for non H2 antigens. J Immunol 1990;145:2123-2131 48. Sherwood RA, Brent L, Rayfield LA: Presentation of alloantigens by host cells. Eur J Immunol 1986;16:569-577 49. McWhinnie DL, Thompson JF, Taylor HM, Chapaman JR, Bolton EM, Carter NP, Wood RFM, Morris PJ: Leukocyte infiltration patterns in renal allografts assessed by immunoperoxidase staining of 245 sequential biopsies. Transplant Proc 1985;17:560-564 50. Hoshinaga K, Mohanakumar T, Goldman MH, Wolfgang TC, Szenpetery S, Lee HM, Lower RR: Clinical significance of in situ detection of T lymphocyte subsets and monocyte/ macrophage lineages in heart allografts. Transplantation 1984; 38:634-641 KEY WORDS * cytotoxic T lymphocytes * microvascular endothelium * alloreactivity * allograft rejection Signal requirements for the generation of CD4+ and CD8+ T-cell responses to human allogeneic microvascular endothelium. R Pardi and J R Bender Downloaded from http://circres.ahajournals.org/ by guest on August 3, 2017 Circ Res. 1991;69:1269-1279 doi: 10.1161/01.RES.69.5.1269 Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1991 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7330. Online ISSN: 1524-4571 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circres.ahajournals.org/content/69/5/1269 Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation Research is online at: http://circres.ahajournals.org//subscriptions/