Immunologic Research 2001;23/1:1–21
T Cell-Mediated Antigen Presentation A Potential Mechanism of Infectious Tolerance
Mark D. Mannie Department of Microbiology and Immunology, East Carolina University School of Medicine, Greenville, NC
Abstract
Key Words
Differentiation of the T cell repertoire and the physiology of T cellmediated antigen presentation are reviewed in relation to mechanisms of self-tolerance. Recent research has indicated that T cell development is a continual process that optimizes partial recognition of self as a homeostatic set-point. Specific T cell antigen recognition of partial agonists is intrinsically linked to expression of class II MHC glycoproteins on T cells. Even ligands that act as TCR antagonists in IL-2 production assays have sufficient agonistic strength to induce expression of class II MHC glycoproteins on T cells. Thus, the intrinsic self-reactivity of the T cell repertoire may promote T-APC activity in vivo and may explain why thymic and peripheral T cells express low but significant levels of class II MHC glycoproteins. T-APC activity induces extensive apoptosis among responder T cells, causes desensitization among surviving responders, and has been implicated in the adoptive transfer of tolerance in the Lewis rat model of experimental autoimmune encephalomyelitis. Overall, these findings support a relationship between the partial recognition of self MHC ligands, expression of class II MHC glycoproteins on mature peripheral T cells, tolerogenic T cell-mediated antigen presentation, and desensitization of pathogenic self-reactive T cells.
T lymphocytes Antigen presentation Immunoregulation Autoimmunity T cell receptor MHC
Introduction Mechanisms of peripheral tolerance are not understood despite extensive study. These regulatory mechanisms are central to our understanding of autoimmune disease, allergy, transplantation rejection, cancer immunology, and vaccination. These regulatory mechanisms in part reflect the function of “regulatory”
Dr. Mark Mannie Department of Microbiology and Immunology East Carolina University School of Medicine Greenville, NC 27858-4354 E-mail:
[email protected]
© 2001 Humana Press Inc. 0257–277X/01/ 23/1:1–21/$15.25
T cells that control the balance between tolerance and immunity. A major goal of contemporary immunological research is to identify the specificity, phenotype, and function of regulatory T cells. Recognition of selfMHC ligands in the thymus is critical for the generation of regulatory T cells that ultimately maintain self-tolerance in peripheral tissues (1–8). The overall design of both thymic selec-
1
tion and the continual peripheral selection of the repertoire is to generate a T cell repertoire that recognizes cognate self-MHC ligands as inefficient ligands (i.e., mixed agonists/ antagonists). These mechanisms include negative and positive selection in the thymus as well as mechanisms of peripheral tolerance and homeostatic growth in the periphery. The intrinsic self-reactivity of the T cell repertoire appears fundamental to the regulation of selftolerance. The thesis presented here is that T cell-mediated antigen presentation links the intrinsic self-reactivity of the repertoire with regulatory mechanisms of active self-tolerance. Quantitative and Qualitative Mechanisms of T Cell Antigen Recognition Like other physiological receptors (9), the TCR transduces both qualitative and quantitative information during interactions with cognate MHC/peptide complexes (10). TCR function may be best described by the distinct parameters of affinity and efficacy. Affinity reflects the quantitative parameters of antigen concentration and receptor occupancy, whereas efficacy reflects the quantitative parameter of intrinsic biological activity (i.e., the continuum between full agonism and full antagonism). Two models of TCR function have been proposed to explain the qualitative recognition of MHC ligands as a spectrum of agonists and antagonists. The kinetic hypothesis explains the qualitative nature of TCR signaling based on differences in the dissociation rate of TCR/MHC-peptide complexes (11,12). To account for antagonistic signaling, this model postulates that the half-life of MHCpeptide/TCR complexes controls the extent of assembly of the tyrosine kinase signaling complex. A short half-life may be insufficient for assembly of this complex, but, nonetheless, may temporarily sequester limiting transduction molecules into short-lived nonproductive complexes. The conformational model in con2
trast predicts that specific interactions between MHC-peptide complexes and TCR may trigger or stabilize a critical conformational alteration in the TCR needed for signal transduction (13–15). According to this model, agonistic ligation of TCR elicits a conformational change in the TCR that triggers receptor dimerization and initiation of signal transduction. The latter model is consistent with a dissociation between affinity and biological activity as has been noted for many receptor systems. For example, G-linked receptors in general bind antagonists with much higher affinity than agonists in the presence of endogenous guanine nucleotides (16,17). Intrinsic to our understanding of immune homeostasis is the assumption that agonistic reactivity to self is required for pathogenesis whereas partial recognition of self is qualitatively insufficient to support pathogenicity regardless of T cell frequencies or concentrations of self-antigen. Intrinsic Self-Reactivity of the T Cell Repertoire: Thymic Selection Thymic T cells exhibiting inefficient recognition of self-MHC ligands (as partial agonists/antagonists) are positively selected to constitute the mature T cell repertoire (18–23) (Fig. 1). Apparently, a range of partial TCR/MHC-ligand interactions constituting partial agonistic signals are sufficient to promote T cell viability and maturation but are insufficient to induce apoptosis. In contrast, T cells that experience strong agonistic interactions with self-MHC ligands are deleted from the repertoire. Negative thymic selection has been studied extensively in models whereby transgenic TCR are expressed together with a transgenic self-antigen in thymic tissue. The efficiency of transgenic TCR interaction with thymic self-antigen is directly related to the degree of apoptosis of transgenic T cells. Transgenic T cells that survive deletion lack overt reactivity to the Mannie
Fig. 1. The Original Qualitative Model of T Cell Antigen Recognition and Thymic Selection (119). This model proposes that T cell antigen recognition is based on two distinct attributes; affinity and efficacy. Affinity represents the attractive physical forces that promote specific interactions between TCR and particular MHC/peptide complexes. Efficacy refers to the ability of a given MHC/peptide complex to trigger biological activity. Affinity and efficacy are each independent quantitative parameters. A full agonist stimulates full T cell activation whereas partial agonists and full antagonists exhibit affinity-based interactions with TCR but vary in the qualities necessary to stimulate biological activity. The model also proposes that thymic selection is guided by the efficacy of interactions between TCR and self MHC ligands. Recognition of self-MHC ligands within a continuum of partial to full agonism elicits negative thymic selection and purges the repertoire of clonotypes capable of autoimmune pathogenesis. Recognition of self-MHC ligands within a continuum of full to partial antagonism elicits positive thymic selection and defines the T cells that ultimately comprise the mature T cell repertoire.
self-antigen in peripheral tissues or in cell culture. These transgenic T cells typically lack or show reduced expression of the respective coreceptor proteins and may also exhibit decreased levels of TCR (24–31). Importantly, quantitative changes in the surface expression of the CD4 or CD8 coreceptors translate into qualitative alterations in antigen responsiveness. Downmodulation of CD8 or neutralization of CD4 or CD8 converts T cell recognition of an agonistic antigen into that of an antagonistic antigen (25,32,33). These studies cannot distinguish whether transgenic T cells are expressed with intrinsic variation in TCR and coreceptor levels or whether transgenic T cells differentiate to a less reactive phenotype during encounters with apoptotic antigen. The observation that thymic double-negative (DN) (i.e., double-positive dull) T cells are enriched for apoptotic cells lends credence to the view that negatively selected thymocytes differentiate by active T Cell-Mediated Antigen Presentation and Tolerance
downregulation of coreceptors (34). Furthermore, negative selection of H-Y reactive T cells was less efficient in bcl-2 transgenic male mice, owing to impaired apoptosis, but, nonetheless, resulted in tolerance to H-Y coupled with the accumulation of anti-H-Y bearing T cells expressing abnormally low levels of CD8 (35). Thus, via a process of TCR or coreceptor downregulation, thymocytes experiencing strong agonistic signaling appear to permanently downmodulate antigen reactivity to avoid activation-induced apoptosis and pathogenic self-reactivity. This process of T cell desensitization may preserve T cells that would otherwise be lost to apoptosis and may enable commitment of these T cells to the positively selected T cell repertoire. Intrinsic Self-Reactivity of the T Cell Repertoire: Peripheral Selection The processes of positive and negative selection that shape the immature T cell reper3
toire in the thymus are paralleled by similar mechanisms that continuously shape the repertoire of mature T cells in peripheral tissues. That is, positively selected T cells that emerge from the thymus continue to interact with positively selecting self-MHC ligands in peripheral tissues. Continuing low efficiency interactions with self appear key for the survival of the mature T cell repertoire (36–39). Likewise, mechanisms of TCR/coreceptor desensitization that downmodulate antigen reactivity in the thymus (40) are mirrored by similar mechanisms that cause desensitization of mature T cells in the periphery. Indeed, T cell selection and differentiation in the thymus and peripheral tissues seem dedicated to establishing nonpathogenic self-reactivity as set-point for selection and a prerequisite for survival. Accordingly, thymic and peripheral mechanisms that couple T cell differentiation and selection ensure T cell recognition of self as a partial agonist/antagonist. Intrinsic Self-Reactivity of the T Cell Repertoire: Positive Selection in the Periphery Immunologists have long held that the T cell repertoire is maintained by homeostatic mechanisms without need for TCR engagement. This model was supported by the observation that stable long-lived populations of naive quiescent T cells persist in thymectomized mice without overt exposure to antigen (41). This view has been modified to accommodate a number of observations. For example, CD4+ T cells require MHC class II glycoproteins for extended in vivo survival (42). Expression of MHC class II glycoproteins in the thymus but not in the periphery supports thymic maturation but is coupled with the gradual disappearance of exported CD4+ T cells (43,44). Similar findings have been noted for persistence of CD8+ T cells in the
4
periphery. Reconstitution of thymectomized male or female recipient mice with bone marrow from anti-H-Y TCR transgenic donors revealed selective expansion of “tolerant” antiH-Y transgenic T cells in males but not in females (45,46). This and other evidence indicate that peripheral maintenance of transgenic T cells requires not only the selecting MHC molecules but also the cognate peptide(s) originally responsible for positive thymic selection. For example, naive polyclonal and transgenic T cells that are transferred into normal mice exhibit little or no cell division but proliferate extensively when transferred into T celldeficient recipients (37). Cell division of donor CD4+ T cells does not occur without the appropriate allelic I-A molecule or in H-2M–/– hosts expressing an overabundance of an appropriate I-A molecule associated with a noncognate peptide (CLIP). However, donor CD4+ T cells originally selected in H-2M–/– donors exhibited extensive proliferation when transferred into T cell-deficient H-2M –/– hosts (37,47). Expansion of donor CD8+ transgenic T cells in lymphopenic recipients required a positively selecting antagonist peptide as a transgene together with expression of the appropriate class I MHC allele (36). Expansion of donor CD8+ T cells in response to the antagonist peptide did not elicit differentiation of cytotoxic T cells. Thus, the antagonist peptide had sufficient agonistic activity in vivo to elicit survival and proliferation at thresholds safely below those necessary for acquisition of effector activity. Overall, these data indicate that TCR-based self-recognition events are critical for the maintenance of the T cell repertoire (48). Furthermore, these studies reveal that a continuum of positive selection events initiated in the thymus continues throughout life in peripheral tissues to continually hone recognition of self as a partially agonistic ligand.
Mannie
Intrinsic Self-Reactivity of the T Cell Repertoire: Negative Selection in the Periphery Likewise, negative thymic selection appears recapitulated in peripheral tissues as a major mechanism of peripheral tolerance. Hence, mature immunocompetent T cells in peripheral tissues that recognize self with high efficiency undergo activation-induced apoptosis but this process is balanced with differentiation to a less-reactive phenotype as a mechanism to avoid self-reactivity (49,50). For example, in double transgenic mice bearing a transgenic Kb-specific TCR and a Kb transgene expressed in extrathymic neuroectodermal tissues, transgenic T cells exhibited normal thymic maturation but were absent from secondary lymphoid organs due to downregulation of both CD8 and TCR (29). Several additional transgenic systems have been used to study TCR and coreceptor downregulation as a mechanism of self-tolerance. These studies have shown that the degree of downregulation is proportional to the antigen load in the periphery. Mice transgenic for Kb exhibited partial downregulation of TCR when exposed to Kb expressed on hepatocytes, whereas exposure to additional sources of Kb resulted in complete TCR downregulation (51,52). H-Y specific CD8+ transgenic T cells transferred into male recipient nude mice expanded vigorously for 5 d but did not cause pathology and largely disappeared after 9 d (50). Surviving transgenic T cells were completely unresponsive upon restimulation with H-Y and exhibited downregulated TCR and CD8. Also, CD8+ transgenic T cells exposed to superantigen in vivo exhibited TCR and coreceptor downregulation together with severely compromised effector function (53,54). The lpr and gld mutant mice also provide evidence for coreceptor downregulation as a mechanism of peripheral tolerance. Deficiency
T Cell-Mediated Antigen Presentation and Tolerance
of fas or fas-ligand in lpr and gld mice results in lymphoproliferation and systemic autoimmunity presumably due to the inability to delete self-reactive T cells in the periphery (55). Thymic selection appears normal in these mice, but defective peripheral regulation among thymus-derived T cells results in a lymphoproliferative syndrome marked by massive lymph node and spleen enlargement and the accumulation of TCRα/β CD4–CD8– T cells. This expanded population of DN T cells appears derived from double positive precursors from the thymus (56). Thus, antigeninduced coreceptor downregulation and activation-induced apoptosis may represent complementary mechanisms of self-tolerance, but the former cannot compensate for the lack of the latter in this particular mouse strain. Indeed, lpr mice overexpressing a bcl-2 transgene have dramatically increased accumulations of DN T cells (57). Consistent with this possibility, anti-H-Y transgenic T cells accumulate in peripheral tissues of male mice despite extensive negative thymic selection, but these transgenic T cells bear a desensitized CD4–CD8– or CD4 –CD8low phenotype and show low levels of “benign” autoreactivity to H-Y (58). Also, mice transgenic for a SV40specific TCR, when expressed on a lpr background and exposed to cognate peptide in vivo, exhibited an accumulation of DN T cells in secondary lymphoid organs (59). This accumulation of DN T cells did not occur in CD95intact lpr/+ mice. T cell desensitization may therefore represent a compensatory mechanism that is most apparent in the absence of apoptosis and may be common to both thymic selection and peripheral tolerance. This mechanism may constrain self-reactivity of both immature and mature T cells to provide a qualitative margin of safety against pathogenic autoreactivity. A diagram portraying selection and differentiation of the T cell repertoire in the thymus and periphery is shown in Fig. 2.
5
Fig. 2. T Cell Recognition of Self-MHC Ligands as Mixed Agonists/Antagonists may Represent a Homeostatic Set-Point that Guides T Cell Differentiation in the Thymus and Periphery. The y-axis portrays clonotypic recognition of cognate self-MHC ligands with increasing efficiency from bottom to top. The x-axis represents milestones of T cell differentiation. As reviewed in the text, immature thymic T cells and mature peripheral T cells exhibit extensive apoptosis when confronted with ubiquitous self-antigens that are perceived as a full agonists. Some of these T cells may also exhibit desensitization coupled with TCR and/or coreceptor downregulation. Permanent desensitization may enable clones to recognize a particular self-MHC ligand as a partial agonist rather than as a full agonist. These T cells may thereby avoid apoptosis and instead may adaptively contribute to the positively selected repertoire. The model incorporates evidence that selfMHC ligands that drive positive thymic selection also represent key survival signals that maintain the mature peripheral repertoire (36,37).
The adaptive advantage of such a repertoire in promoting crossreactive recognition of foreign antigen is shown in Fig. 3. Partial Agonism, Anergy, and Infectious Tolerance Partial agonists/antagonists are usually studied as synthetic peptides with substitutions in particular T cell contact residues. However, a more physiological system is represented by self-antigens serving as partial agonist/antagonists and analog peptides as fully agonistic foreign antigens. Partial agonists effectively stimulate certain responses in cloned T cell lines but do not elicit the com-
6
plete repertoire of responses elicited by full agonists (60–62). “Full” TCR antagonists may not exhibit stimulatory activity at any concentration for a given response. Rather, at higher concentrations, antagonistic ligands simply exhibit more complete inhibition of an agonistic response. However, “full” TCR antagonists usually exhibit properties of a mixed agonist/antagonist when tested in more sensitive assays. Most biologically active antigens appear to have reciprocal degrees of agonistic or antagonistic activity. For example, a series of altered peptide ligands based on the LCMV glycoprotein exhibited activities ranging from a strong agonist to no activity and conversely the opposite spectrum of activity
Mannie
Fig. 3. A T Cell Repertoire Optimized to Recognize Self as a Partial Agonist is Primed to Recognize Foreign Peptides as Full Agonists. The positivelyselected repertoire recognizes self-peptide/self-MHC complexes as specific, nonactivating ligands. These TCR/MHC interactions thereby exhibit affinity but lack efficacy. Whereas a nonselected repertoire would have to crossreact in regard to both affinity and efficacy, a positively-selected repertoire has already ensured affinity for self-MHC and would require crossreactivity only in regard to efficacy. An extension of this model addresses mechanisms by which T cells specific for foreign peptides versus those that agonistically recognize self are able to mediate adaptive immunity or are rendered tolerant respectively in the same location at the same time (120).
ranging from no antagonism to strong antagonistic activity, respectively (63). Partial agonists that have sufficient agonistic strength appear to have important tolerogenic activities. For example, some partial agonists/antagonists induce anergy in vitro (64,65). Administration of certain partial agonists in rodent models of autoimmune disease also prevents or inhibits pathogenesis (66–68). Expression of an antagonistic peptide as a transgene strongly inhibited agonist-driven responses of transgenic T cells in vivo (69). The ability of partial agonists to generate anergy may be important because anergy has been linked to mechanisms of infectious tol-
T Cell-Mediated Antigen Presentation and Tolerance
erance (i.e., active suppression). In one experimental model, for example, transgenic T cells specific for the S1 determinant of influenza hemagglutinin were abundant in hosts bearing the viral glycoprotein as a transgenic selfmolecule. These self-reactive transgenic T cells however lacked responsiveness to the viral self-antigen. Analysis of T cell populations revealed an anergic population of CD25hi, CD45RBint transgenic T cells as well as a separate subset of fully reactive T cells. The anergic T cells suppressed the activity of the latter T cells by an undefined mechanism (70). Recognition of self as a partial agonist may thereby promote “regulatory” activity by T cells. Thymectomy of neonates in some mouse strains causes a spectrum of organ-specific autoimmune diseases (2,3). Subsets of CD4+ T cells are responsible for both autoimmune pathology and for prevention of autoimmunity. The latter population of regulatory T cells uniquely bear CD25, appear anergic to stimulation through the TCR (i.e., recognize antigen as a partial or inefficient ligand), suppress the proliferation of other CD4+ T cells, and mediate active suppression of autoimmunity upon adoptive transfer into thymectomized recipients. These regulatory T cells also suppress autoimmunity when transferred into recipients of cloned autoreactive T cells. The TCR specificity of these CD25+ regulatory T cells appears critical for suppressive activity because CD25+ thymocytes from TCR transgenic mice do not express regulatory activity. CD25 expression, however, is not correlated with suppressive function because activation of CD4+CD25– T cells results in ample surface CD25 without expression of regulatory activity. Hence, positive selection in the thymus results in the export of regulatory T cells that control the differentiation of pathogenic selfreactive T cells in peripheral tissues.
7
T Cell-Mediated Antigen Presentation and Tolerance Induction Mouse (71–81), rat (82–88), and human (89–105) T cells are known to express MHC class II glycoproteins under conditions of inflammation, activation, or transformation. T cell expression of MHC class II glycoproteins serves the classical function of antigen presentation. Rested lines of T cells in vitro are devoid of MHC class II glycoproteins, but these proteins are expressed during activation with antigen and professional APC. Unlike professional APC, antigen presentation by T-APC is often inhibitory and results in apoptosis or anergy of T cells. A number of studies have collectively revealed the tolerogenic influence of T cell-mediated antigen presentation (101–108). Indeed, the prototypic experimental evidence for MHC-restriction in an in vitro model of tolerance (anergy) was based on a human model of T cell-mediated antigen presentation (94,109). Our interest in T cell-mediated antigen presentation stemmed from studies of in vitro anergy (110,111). In these experiments, myelin basic protein (MBP)-specific T cells were cultured with antigen and splenic APC in the presence or absence of tolerogenic antiCD4/anti-LFA-1 mAb. T cells recovered from these initial cultures were propagated for 1–2 wk, and those initially exposed to tolerogenic mAb were found to lack responsiveness when optimally restimulated with splenic APC and antigen. After removal from the tolerogenic mAb immediately after the initial 3-d culture, T cells paradoxically exhibited a re-bound proliferative response that preceded induction of long-lasting anergy. The proliferative burst was associated with the expression of class II MHC glycoproteins on T cells, was not altered by addition of exogenous antigen, and was blocked by mAb against I-A, CD4, or LFA-1. Thus, exposure to tolerogenic mAb in the presence of antigen and splenic 8
APC was responsible for the expression of I-A glycoproteins on T cells, the proliferative burst after removal of tolerogenic mAb, and the subsequent induction of anergy. Similar experiments were performed in which irradiated splenic APC were pulsed with antigen so that the only source of antigen was that associated with professional APC in the initial culture. Here again, tolerogenic mAb inhibited antigen reactivity in the initial culture but elicited T cell expression of I-A and rebound proliferation in the second culture as T cells emerged from the mAb-induced blockade. Our interpretation of these data was influenced by earlier observations that class II MHC glycoproteins were shed by professional APC and subsequently acquired by responder T cells (76,80,100). That is, MBP-specific T cells appeared to acquire MBP-derived I-A complexes from professional APC during the initial culture, and in the presence of tolerogenic mAb, retained these I-A complexes so that in the second culture, these I-A+ T cells reciprocally presented MBP-derived I-A complexes after removal of the tolerogenic mAb. Importantly, these I-A+ MBP-specific T cells also mediated adoptive transfer of tolerance in Lewis rat recipients challenged ~2 wk later with an encephalitogenic dose of MBP in CFA. Overall, this research revealed a relation between a partial blockade of antigen recognition via anti-CD4/anti-LFA-1 mAb, a phase of T cell-mediated antigen presentation, induction of in vitro anergy, and adoptive transfer of tolerance to EAE. CD4 Controls the Biological Efficacy of T Cell Antigen Recognition Although the above studies revealed that anti-CD4 and anti-LFA-1 mAb were synergistic for promotion of T-APC activity, a comparison of the independent activities of anti-CD4 versus anti-LFA-1 mAb revealed that anti-CD4 mAb were more effective for genMannie
eration of T-APC activity (111). In this MBPspecific clone, anti-CD4 mAb did not prevent antigen recognition per se but rather converted recognition of autologous rat (R) MBP from that of a partial agonist into that of a full antagonist (32). In the presence of anti-CD4 mAb, guinea pig (GP) MBP-stimulated responses were half-maximally inhibited by equimolar concentrations of rat (R)MBP. However, ConA stimulated responses were not affected by RMBP. The ability of anti-CD4 mAb to convert recognition of an agonistic antigen into that of an antagonist was also observed in polyclonal lines and numerous other CD4+ clones (112). The findings also appeared applicable to the CD8+ T cell lineage, because an antiCD8 mAb converted recognition of an agonistic antigen by CD8+ transgenic T cells into that of a specific TCR antagonist (25). Overall, these data indicate that coreceptors may determine T cell perceptions of agonism vs antagonism. T Cell-Mediated Antigen Presentation and Infectious Tolerance In Vitro The sole structural difference distinguishing GPMBP and RMBP in the dominant encephalitogenic 72–86 sequence was a S 80/T 80 substitution. A majority of MBPspecific T cell clones (i.e., the GP2 set of clones) react more potently to GPMBP than RMBP, recognize RMBP as an antagonist in the presence anti-CD4 mAb, and when activated, mediate mild or no EAE (32,111). Another set of clones (i.e., the R2 set of clones) do not recognize position 80 as a T cell contact residue and thereby recognize GPMBP and RMBP with high levels of equipotent reactivity. When activated, R2 T cells express potent encephalitogenic activity. GP2 T cells that were cultured for 3 d with GPMBP, irradiated APC, and anti-CD4/LFA-1 tolerogenic mAb were found after recovery from this culture to express high levels of I-A glycoproT Cell-Mediated Antigen Presentation and Tolerance
teins. These I-A+ GP2 T cells mediated T-APC activity because these T cells presented MBP to R2 responders. In these assays, the only source of antigen was added as a pulse of irradiated splenic APC or was added as soluble antigen to the initial culture. After acquisition of GPMBP/ I-A complexes, GP2 T cells were purified on discontinuous gradients of Percoll and propagated for 1–3 d in IL-2, were irradiated, and were used as T-APC to stimulate R2 responders. GP2 T cells loaded with RMBP exhibited long-lasting T-APC activity that persisted for a week of propagation in IL-2, whereas T cells loaded with GPMBP exhibited T-APC activity for 1–3 d after the initial culture (111). Presentation of RMBP by irradiated GP2 T-APC resulted in the proliferation of R2 T cells, but, subsequently, these activated R2 T cells exhibited anergy. Thus, in the presence of antigen, tolerogenic mAb induced T-APC activity and eventual anergy in GP2 T cells, and the anergic fate of GP2 T cells was transferred to responder R2 T cells by a mechanism of T cell-mediated antigen presentation. Partial T Cell Antigen Recognition Confers Prolonged T-APC Activity Experiments with anti-CD4 mAb and comparisons of GPMBP and RMBP implicated partial or antagonistic antigen recognition in the generation of persistent T-APC activity. That is, RMBP was more potent than GPMBP for generation of T-APC activity in GP2 clones even though GPMBP was a more potent antigen. Furthermore, anti-CD4 mAb were critical to our discovery of T-APC activity. These observations were paradoxical because strong antigens would predictably be superior for induction of activation molecules such as I-A on T cells. To assess whether anti-CD4 mAb promoted T-APC activity by neutralization of CD4 or by negative signaling, a set of CD4+ and CD4– 9
clones were isolated that were specific for the minor 55–69 or the dominant 72–86 sequence of GPMBP (112). Expectedly, T cells specific for the GP55–69 region exhibited approx 10-fold less reactivity to GPMBP than GP72–86 specific clones. GP55–69 specific clones did not recognize RMBP as an agonist or antagonist due to substantial sequence differences. CD4– GP55–69 reactive clones that were activated with GPMBP and irradiated APC subsequently mediated superior T-APC activity compared to CD4+ GP55–69 specific clones. Conversely, CD4– and CD4+ clones specific for GP72–86 did not exhibit detectable T-APC activity after activation with GPMBP. When GP72–86 specific clones were stimulated with the partial agonist RMBP however, CD4– clones exhibited high levels of T-APC activity whereas CD4+ clones lacked T-APC activity. Tolerogenic anti-CD4/ LFA-1 mAb enabled expression of T-APC activity when GP72–86 specific CD4+ T cells were cultured with RMBP. T-APC activity of CD4– clones was essentially equivalent to CD4+ clones that were cultured with tolerogenic mAb. Thus, anti-CD4 mAb augmented T-APC activity by neutralization of CD4. These studies showed that T cells with potent reactivity to GPMBP or RMBP subsequently lacked the ability to present the respective antigen, whereas T cells exhibiting partial or inefficient reactivity were highly effective T-APC for presentation of that antigen. These findings indicated that the efficacy of antigen recognition was inversely related to the expression of T-APC activity. Agonistic Recognition of MHC Ligands on T Cells Causes Accelerated Loss of T-APC Activity The finding that tolerogenic mAb and partial antigen recognition enabled T-APC activity suggested that either partial TCR/MHC
10
interactions promoted T-APC activity or fully agonistic MHC/TCR interactions inhibited T-APC activity. The most revealing observation was that partial agonists were superior to full agonists as measured by persistence of I-A on responding T cells. Also, both full and partial agonists elicited I-A/MBP complexes on responder T cells, but subsequent recognition of full agonists presented by T cells accelerated the clearance of I-A molecules from the T cell surface (113). Several lines of evidence supported this scenario (113). First, tolerogenic mAb that strongly promoted expression of T-APC activity during a 3-d culture with irradiated APC and MBP also maintained T-APC activity of purified T cells when included in the subsequent 4-d culture in IL-2. The absence of tolerogenic mAb in either the first or second culture resulted in the virtually complete loss of T-APC activity. Second, tolerogenic mAb preserved T-APC activity initiated by an antigen pulse in cultures of transformed, MBPspecific T cells that constitutively synthesized class II MHC glycoproteins. The latter experiments were performed in a pure, rapidly dividing T cell clone in the absence of professional APC. Third, the best predictor of strong T-APC activity was inversely related to the ability of T cells to respond to cognate antigen presented by T cells of the same clone. Fourth, T-APC that were loaded with both MBP and conalbumin that were cultured with irradiated MBP-specific responders lost the ability to present MBP in a subsequent culture but retained the ability to present conalbumin. The same T-APC, when cultured with irradiated conalbumin-specific responders, lost the ability to present conalbumin but not MBP. These observations are most consistent with the possibility that tolerogenic mAb acted on T-APC to preserve I-A/antigen complexes by blocking cognate T-APC/responder inter-
Mannie
actions. Overall, these findings provided a mechanistic basis to explain the superior activity of partial agonists for expression of T-APC activity. Partial Agonists have Sufficient Activity to Stimulate Acquisition of MHC/Peptide Complexes GP2 T cells cultured with irradiated APC that were pulsed with both GPMBP and conalbumin in the presence of tolerogenic mAb presented each antigen to the respective responders in a subsequent assay (112–114). T cell acquisition of the unrelated antigen conalbumin from professional APC required both syngeneic APC and recognition of MBP. An anti-I-A mAb that inhibited recognition of MBP blocked acquisition of MBP-derived I-A complexes by responder T cells but only when the anti-I-A mAb was added at the initiation of culture (114). Addition of the antiI-A mAb after the first day of a 3-d culture had no effect; hence, the inhibitory activity of anti-I-A mAb was not attributed to carryover of mAb into the T-APC/responder assay. TCR ligation was also required for antigen acquisition by transformed T cells that constitutively synthesized I-A glycoproteins. That is, transformed T cells acquired MBP from antigen-pulsed syngeneic APC but did not acquire antigen from antigen-pulsed allogeneic SPL or from syngeneic SPL in the presence of anti-I-A mAb. Nonetheless, these transformed T cells readily acquired soluble antigen added to the culture for subsequent presentation. Hence, antigen recognition occurred in the presence of anti-CD4/LFA-1 mAb and was necessary for acquisition of MHC/peptide complexes from antigen-pulsed professional APC. The requirement for antigen ligation even in the presence of anti-CD4/LFA-1 mAb sug-
T Cell-Mediated Antigen Presentation and Tolerance
gested that even partial MHC ligands were sufficient to catalyze antigen acquisition by T cells. To address this issue, we focused on a GP2 subclone (3H3) that exhibited a desensitized phenotype (i.e., CD4 –CD8 –CD5 low TCRlow). This clone recognized GPMBP as an agonist but recognized RMBP as a TCR antagonist even in the absence of anti-CD4 mAb (115). RMBP behaved as a full TCR antagonist and competitively inhibited GPMBPstimulated proliferation and IL-2 production but did not affect Con A-stimulated responses. RMBP was also an antagonist of GPMBPstimulated induction of B7.1 costimulatory molecules on T cells. Even though RMBP was a TCR antagonist in these assays, RMBP exhibited sufficient agonistic activity to efficiently stimulate acquisition of RMBP/MHC complexes by 3H3 T cells for subsequent antigen presentation. Overall, RMBP exhibited a hierarchical induction of APC-associated molecules on 3H3 T cells (I-A > B7.2 > B7.1), whereas GPMBP maximally induced all three. RMBP antagonized GPMBP-stimulated responses according to the reverse rank order (B7.1 > B7.2 > I-A). Thus, recognition of autologous RMBP on professional APC elicited the accumulation of RMBP-derived I-A complexes on 3H3 T cells but actively antagonized expression of B7.1 (115). 3H3 T cells that were cultured with irradiated splenic APC and RMBP were tolerogenic in vivo in that these 3H3 T-APC rendered Lewis rat recipients resistant to the subsequent active induction of EAE (unpublished data). Purified MBP-Pulsed T-APC Mediate Adoptive Transfer of Tolerance in EAE We have isolated several transformed MBPspecific T cell subclones that constitutively express class II MHC glycoproteins (113,114,116). When activated, normal T cell
11
clones undergo blastogenesis, upregulate IL-2 receptors, and exhibit rapid IL-2dependent proliferation. During the next 1–2 wk, T cells progressively downregulate IL-2 receptor levels and revert to a small quiescent phenotype. During repeated activation/rest cycles of in vitro propagation, transformed variants infrequently arise that retain a blastogenic phenotype and do not revert to a rested phenotype. These variants easily overgrow normal T cells in the culture. These blastogenic variants express high levels of class II MHC glycoproteins, B7.1/B7.2 costimulatory molecules, and IL-2 receptors and exhibit rapid IL-2-dependent growth. These clones also retain normal levels of TCR and CD4 and directly respond to cognate antigen in the absence of professional APC. When pulsed with MBP, these transformed T cells present antigen and cause activation of responder T cells. A pulse with MBP also enables these transformed T cells to mediate adoptive transfer of tolerance in the Lewis rat model of EAE (116). Thus, in three experimental systems, T-APC activity was associated with adoptive transfer of tolerance. Culture conditions optimal for expression of T-APC activity were also optimal for adoptive transfer of tolerance. For example, the CD4+ GP2 T cell clone required culture with RMBP, irradiated splenic APC, and tolerogenic mAb to express high levels of T-APC activity and to mediate adoptive transfer of tolerance (111). The CD4– 3H3 T cell clone also required culture with RMBP and irradiated splenic APC but did not require tolerogenic mAb to generate optimal T-APC activity or to transfer tolerance (115) (unpublished data). However, interpretation of these findings must accommodate the possibility that residual professional APC in the T-APC preparation may influence tolerogenic activity. In contrast, transformed T-APC only required a short pulse with MBP to mediate T-APC activity and these conditions were suf-
12
ficient for adoptive transfer of tolerance (116). The latter observation discounted the possibility that residual professional APC in T-APC preparations were required for induction of tolerance in vivo. T Cell Acquisition of Class II MHC Glycoproteins Constitutive synthesis and expression of class II MHC molecules by transformed T cells provides suggestive evidence that normal activated T cells may also have the capacity to synthesize and express class II MHC glycoproteins. However, normal T cell lines that were activated in the absence of APC by various mitogens exhibited blastogenesis but did not express I-A. These findings revealed a fundamental difference between transformed T-APC that constitutively synthesize I-A and normal activated T cell blasts that express no detectable surface I-A unless cultured with a source of I-A+ APC. Even I-A+ transformed T cells needed cognate antigen recognition to acquire MBP/I-A complexes from MBP-pulsed professional APC (114). In contrast, these T-APC did not need TCR ligation to acquire antigen for presentation when soluble MBP was added to the culture. These findings indicated that transformed T-APC may acquire I-A/peptide complexes via two distinct mechanisms: one that involved acquisition of preformed I-A/peptide complexes shed from professional APC and a separate mechanism involving direct biosynthesis of I-A and assembly of I-A/peptide complexes. Both mechanisms enable T cells to present antigens available in the immediate environment. Of the two mechanisms, direct acquisition of I-A/peptide complexes appeared to be dominant in controlling expression of I-A on normal MBP-specific T cells in vitro. In cultures comprised of transformed I-A+ T-APC and MBP-specific T cell responders, addition
Mannie
of MBP elicited the rapid expression of I-A on responders (116). Indeed, TCR ligation elicited I-A expression on responder T cells within 1 h of antigen exposure by a mechanism that was resistant to the protein synthesis inhibitor cycloheximide. The source of I-A on responders appeared to derive from T-APC because MBP also elicited acquisition of PKH26-labeled membrane from prelabeled T-APC. Also, MBP-specific Brown Norway (BN) rat T cells acquired Lewis rat I-A molecules when activated with F1(BN × Le) APC and MBP. Several antigen-specific T cell lines were tested in the presence of antigen, and all efficiently acquired I-A molecules from a variety of I-A + APC, including T-APC, transformed MØ, and irradiated splenocytes. Hence, mechanisms mediating the intercellular transfer of I-A molecules appeared relevant for a wide array of responders and APC. The finding that specific antigen recognition elicited intercellular I-A acquisition was consistent with two possibilities. TCR ligation may trigger activation-dependent processes that independently mediate acquisition of APC-derived I-A molecules, or alternatively, the TCR may serve as a docking molecule that strips APC-derived vesicles for ingestion by responding T cells. Several observations were consistent with the former mechanism (116). First, T cells that were preactivated with PMA and ionomycin and then cultured with T-APC acquired essentially the same level of APC-derived I-A regardless of the presence or absence of specific antigen. Second, MBP-specific BN T cells efficiently acquired I-A from completely allogeneic Lewis splenocytes when activated by Con A. Third, Balb/c thymocytes that were preactivated with ionomycin and PMA also acquired xenogeneic I-A from Lewis rat T-APC together with rat IL-2 receptor and rat CD4 (116). Fourth, in 4-h cultures of rested MBP-specific responders and I-A + T-APC, MBP caused
T Cell-Mediated Antigen Presentation and Tolerance
accumulation of I-A on responders by a mechanism that was blocked by neutralizing mAb against LFA-1, CD4, I-A, or TCR. In contrast, I-A acquisition by preactivated responders was not inhibited by any of these mAb either alone or together (unpublished data). The mechanism responsible for the intercellular transfer of I-A is not currently established. Rat I-A+ T-APC used in these studies expressed abundant class II MHC glycoproteins in multivesicular MHC class II-enriched compartments (MIIC). Vesicles (i.e., referred to as exosomes) within the MIIC compartment express high levels of class II MHC/peptide complexes. These vesicles are released into extracellular space when the MIIC fuses with the APC surface (117). I-A+ rat T-APC also release class II MHC-bearing vesicles (118). When T-APC were cultured with MBP, the T-APC-derived vesicles directly activated MBP responders. Activated responders directly acquired vesicular I-A/MBP complexes and presented these complexes to activate a second set of responders. Because these studies were initiated with rested responders, MHC-restricted antigen recognition was required for acquisition of I-A/MBP complexes. These data implicate vesicular origins of I-A/peptide complexes in the intercellular transfer of I-A from APC to responders. T Cells may Compete for the Acquisition of Self-MHC Ligands Professional APC process and present selfMHC ligands as a normal process of immune homeostasis. These self-MHC ligands would be predominantly recognized as partial ligands by the positively selected repertoire. Partial interactions between self-MHC ligands and TCR appear sufficient for T cell acquisition of these self-MHC complexes (115). Thus, normal T cells of the repertoire may acquire and present MHC complexes comprised of cognate
13
Fig. 4. MHC/Antigen Complexes are Shed from Professional APC and Accumulate on Responder T Cells. Professional APC shed MHC/antigen complexes that are acquired and expressed by responder T cells. Even MHC/peptide complexes that are full TCR antagonists in IL-2 production assays have sufficient agonistic activity to elicit accumulation of MHC complexes on responder T cells (115). Normal T cells of the positively-selected repertoire may therefore recognize cognate self-MHC ligands in peripheral tissues with sufficient agonism to acquire those self-MHC ligands for presentation. These T cells would present these selfMHC ligands only to T cells that recognize those self-MHC ligands as full agonists. During T-APC/responder interactions, pathogenic responders expressing the highest levels of autoreactivity would most successfully compete for these self-MHC ligands and then would reciprocally present these self-antigens among each other.
and bystander peptides as guided by clonotypic TCR specificity. In support, normal thymic and splenic T cells express low but significant levels of class II MHC glycoproteins (116). Self-peptides presented by positively selected T cells would primarily target T cells that escaped negative thymic selection and that express high levels of pathogenic self-reactivity. These pathogenic T cells would in turn capture these self-MHC complexes to mediate reciprocal antigen presentation. According to this model, self-MHC ligands may be transferred from professional APC to positively selected T cells and from these T-APC to pathogenic responders until these self-MHC ligands are concentrated onto the T cells exhibiting the strongest self-reactivity to that peptide (Fig. 4). T-APC Cause T Cell Desensitization In Vitro Transformed I-A+ T-APC were tolerogenic in vivo (116). These T cells were likewise highly 14
tolerogenic in vitro (unpublished data). MBPspecific T-APC/responder interactions resulted in the virtual elimination of T-APC within 1 d of culture and extensive death of responders 3–4 d after initiation of culture. The functional distinction between T-APC and responders was quickly lost in these cultures because responders acquired I-A/peptide complexes and mediated reciprocal antigen presentation. Responders that survived this culture exhibited losses of antigenic responsiveness that remained profound for more than 1 mo of propagation in IL-2. Also, antigen-specific T-APC/responder interactions resulted in outgrowth of desensitized CD4– variants. Presumably, responders that downregulated CD4 were less susceptible to activation-induced cell death and preferentially expanded in IL-2. High concentrations of MBP also caused extensive death of I-A+ transformed T-APC in the absence of responders. Here again, exposure to antigen resulted in the preferential IL-2 dependent outMannie
Fig. 5. The Ability of T-APC to cause Desensitization of Responders may be part of a Regulatory Loop for the Active Maintenance of Self-Tolerance. This model is based on the observations that partial agonists are superior to full agonists for generation of T-APC activity and that T-APC cause desensitization of responders. Desensitization by definition affects the efficiency of antigen recognition such that desensitized responders recognize the given self-MHC ligand as a mixed agonist/antagonist. A regulatory loop may exist in which T-APC activity of the positively-selected repertoire would cause desensitization of pathogenic responders that in turn would lose pathogenic activity and contribute to T-APC activity to cause desensitization of additional pathogenic responders. This mechanism may constitute a mechanism of infectious tolerance.
growth of CD4– T-APC. Repeated exposures to antigen resulted in the complete overgrowth of the T-APC by CD4– variants. Thus, reciprocal antigen presentation among CD4+ T-APC results in the preferential IL-2-dependent outgrowth of CD4– T-APC. In accordance with previous research, CD4– T cells would predictably exhibit an approximate 100-fold loss in antigenic reactivity and would exhibit superior T-APC activity. Overall, these findings reveal a potential regulatory loop. Antigen presentation by T-APC causes long-term desensitization of responders including outgrowth of CD4– variants. Desensitization of responders impairs the efficiency of antigen recognition and thereby promotes expression of T-APC activity but decreases expression of pathogenic activity. Desensitized T cells prone to mediate presentation of a self-antigen in turn may cause desensitization of additional responders. In the presence of antigen, continuing T-APC activT Cell-Mediated Antigen Presentation and Tolerance
ity coupled with continued desensitization of responders would progressively expand the pool of T cells mediating tolerogenic T-APC activity while depleting the pool of T cells capable of effector action against the selfantigen. This regulatory loop may represent a mechanism of infectious tolerance that continually amplifies in the presence of selfantigens (Fig. 5). Summary Here, we review the experimental foundations of a model that couples the inherent selfreactivity of the T cell repertoire, T-APC activity, and infectious tolerance. Although we study this model in regard to autoreactivity against MBP, this model (Fig. 6) would be applicable to self-antigens in general. The basic tenets of the model (Fig. 6) are as follows: Mechanisms of thymic and peripheral selection promote the differentiation of 15
Fig. 6. Selection in the Thymus and Periphery Ensure High Frequencies of T-APC and Relatively Low Frequencies of Pathogenic Responders. In regard to a given self antigen (e.g., MBP), relevant T cells that recognize MBP or a crossreactive self-peptide as a mixed agonist/antagonist would exist at higher frequencies than pathogenic T cells that recognize MBP as a full agonist. Positively-selected T cells that recognize MBP as a partial agonist would acquire and present these complexes to encephalitogenic T cells and thereby would cause desensitization of these pathogenic T cells. Desensitization of antigen reactivity would abrogate pathogenic activity and would promote presentation of MBP to cause desensitization of additional encephalitogenic T cells.
T cells that recognize MBP-derived I-A complexes (or crossreactive complexes) as partial agonists. Partial reactivity against MBP is not sufficient to support encephalitogenic activity. Partial recognition of these complexes as presented on professional APC is sufficient to induce acquisition of these complexes by these partially reactive MBP-specific T cells. These T cells in turn present MBP-derived I-A complexes to encephalitogenic T cells that (by definition) recognize MBP as a full agonist. Owing to ongoing selection in the thymus and periphery, the majority of MBP-specific T cells would recognize MBP as a partial ligand, whereas only low frequencies of T cells would exhibit pathogenic autoreactivity. T cellmediated presentation of MBP-derived I-A complexes would cause desensitization of
16
encephalitogenic T cells. Desensitization of encephalitogenic T cells would entail differentiation to a less reactive phenotype and would thereby engender recognition of MBP as a partial agonist. Desensitized MBPreactive T cells would thereby lack pathogenic activity and would contribute to ongoing T-APC activity to desensitize additional encephalitogenic T cells. In summary, this model may constitute a simple homeostatic system of infectious tolerance in which regulatory T cell function represents a constitutive function of normal T cells. Acknowledgment This work was supported by a research grant from the National Multiple Sclerosis Society.
Mannie
References 1 Sakaguchi, S, Sakaguchi, N: Thymus and autoimmunity: capacity of the normal thymus to produce pathogenic self-reactive T cells and conditions required for their induction of autoimmune disease. J Exp Med 1990;172: 537–545. 2 Itoh, M, Takahashi, T, Sakaguchi, N, Kuniyasu, Y, Shimizu, J, Otsuka, F, Sakaguchi, S: Thymus and autoimmunity: production of CD25+CD4+ naturally anergic and suppressive T cells as a key function of the thymus in maintaining immunologic self-tolerance. J Immunol 1999;162:5317–5326. 3 Suri-Payer, E, Amar, AZ, Thornton, AM, Shevach, EM: CD4 + CD25 + T cells inhibit both the induction and effector function of autoreactive T cells and represent a unique lineage of immunoregulatory cells. J Immunol 1998;160: 1212–1218. 4 Modigliani, Y, Coutinho, A, Pereira, P, Le Douarin, N, ThomasVaslin, V, Burlen-Defranoux, O, Salaun, J, Bandeira, A: Establishment of tissue-specific tolerance is driven by regulatory T cells selected by thymic epithelium. Eur J Immunol 1996;26:1807–1815. 5 Le Douarin, N, Corbel, C, Bandeira, A, Thomas-Vaslin, V, Modigliani, Y, Coutinho, A, Salaun, J: Evidence for a thymus-dependent form of tolerance that is not based on elimination or anergy of reactive T cells. Immunol Rev 1996; 149:35–53. 6 Wilson, DB, Wilson, DH, Schroder, K: Acquired thymic tolerance and experimental allergic encephalomyelitis in the rat. I. Parameters and analysis of possible mechanisms. Eur J Immunol 1998;28:2770–2779. 7 Khoury, SJ, Sayegh, MH, Hancock, WW, Gallon, L, Carpenter, CB, Weiner, HL: Acquired tolerance to experimental autoimmune encephalomyelitis by intrathymic injection of myelin basic protein or its major encephalitogenic peptide. J Exp Med 1993;178: 559–566.
T Cell-Mediated Antigen Presentation and Tolerance
8 Onodera, K, Volk, HD, Ritter, T, Kupiec-Weglinski, JW: Thymus requirement and antigen dependency in the “infectious” tolerance pathway in transplant recipients. J Immunol 1998;160:5765–5772. 9 Stephenson, RP:A modification of receptor theory. Brit J Pharmacol 1956;11:379. 10 Madrenas, J, Germain, RN: Variant TCR ligands: new insights into the molecular basis of antigendependent signal transduction and T-cell activation. Sem Immunol 1996;8:83–101. 11 Kersh, GJ, Kersh, EN, Fremont, DH, Allen, PM: High- and lowpotency ligands with similar affinities for the TCR: the importance of kinetics in TCR signaling. Immunity 1998;9:817–826. 12 Lanzavecchia, A, Lezzi, G, Viola, A: From TCR engagement to T cell activation: a kinetic view of T cell behavior. Cell 1999;96:1–4. 13 Janeway, CA, Jr., Dianzani, U, Portoles, P, Rath, S, Reich, EP, Rojo, J, Yagi, J, Murphy, DB: Crosslinking and conformational change in T-cell receptors: role in activation and in repertoire selection. Cold Spring Harb Symp Quant Biol 1989;54:657–666. 14 Janeway, CA, Jr., Bottomly, K: Responses of T cells to ligands for the T-cell receptor. Semin Immunol 1996;8:108–115. 15 Alam, SM, Davies, GM, Lin, CM, Zal, T, Nasholds, W, Jameson, SC, Hogquist, KA, Gascoigne, NR, Travers, PJ: Qualitative and quantitative differences in T cell receptor binding of agonist and antagonist ligands. Immunity 1999;10:227–237. 16 Ott, S, Costa, T: Guanine nucleotide-mediated inhibition of opioid agonist binding. Modulatory effects of ions and of receptor occupancy. Biochem Pharmacol 1989;38:1931–1939. 17 Sibley, DR, Mahan, LC, Creese, I: Dopamine receptor binding on intact cells. Absence of a highaffinity agonist-receptor binding state. Mol Pharmacol 1983;23: 295–302.
18 Hogquist, KA, Jameson, SC, Heath, WR, Howard, JL, Bevan, MJ, Carbone, FR: T cell receptor antagonist peptides induce positive selection. Cell 1994;76: 17–27. 19 Saito, T,Watanabe, N: Positive and negative thymocyte selection. Crit Rev Immunol 1998;18:359–370. 20 Teh, HS, Motyka, B, Teh, SJ: Influence of the affinity of selecting ligands on T cell positive and negative selection and the functional maturity of the positively selected T cells. Crit Rev Immunol 1997;17:399–410. 21 Williams, O, Tanaka, Y, Tarazona, R, Kioussis, D: The agonist-antagonist balance in positive selection. Immunol Today 1997;18:121–126. 22 Hogquist, KA, Bevan, MJ: The nature of the peptide/MHC ligand involved in positive selection. Semin Immunol 1996;8:63–68. 23 Goldrath, AW, Bevan, MJ: Selecting and maintaining a diverse T-cell repertoire. Nature 1999;402: 255–262. 24 Kisielow, P, Bluthmann, H, Staerz, UD, Steinmetz, M, von Boehmer, H: Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4+8+ thymocytes. Nature 1988;333:742–746. 25 Jameson, SC, Hogquist, KA, Bevan, MJ: Specificity and flexibility in thymic selection. Nature 1994;369:750–752. 26 Auphan, N, Jezo-Bremond, A, Schonrich, G, Hammerling, G, Arnold, B, Malissen, B, SchmittVerhulst,AM: Threshold tolerance in H-2Kb-specific TCR transgenic mice expressing mutant H-2Kb: conversion of helper-independent to helper-dependent CTL. Int Immunol 1992;4:1419–1428. 27 Husbands, SD, Schonrich, G, Arnold, B, Chandler, PR, Simpson, E, Philpott, KL, Tomlinson, P, O’Reilly, L, Cooke,A, Mellor,AL: Expression of major histocompatibility complex class I antigens at low levels in the thymus induces T cell tolerance via a nondeletional mechanism. Eur J Immunol 1992;22:2655–2661.
17
28 Auphan, N, Schonrich, G, Malissen, M, Barad, M, Hammerling, G, Arnold, B, Malissen, B, Schmitt-Verhulst, AM: Influence of antigen density on degree of clonal deletion in T cell receptor transgenic mice. Int Immunol 1992;4:541–547. 29 Schonrich, G, Kalinke, U, Momburg, F, Malissen, M, SchmittVerhulst, AM, Malissen, B, Hammerling, GJ, Arnold, B: Down-regulation of T cell receptors on self-reactive T cells as a novel mechanism for extrathymic tolerance induction. Cell 1991;65: 293–304. 30 Teh, HS, Kishi, H, Scott, B, Von Boehmer, H: Deletion of autospecific T cells in T cell receptor (TCR) transgenic mice spares cells with normal TCR levels and low levels of CD8 molecules. J Exp Med 1989;169:795–806. 31 Budd, RC, Mixter, PF: The origin of CD4 – CD8 – TCRαβ + thymocytes: a model based on T-cell receptor avidity. Immunol Today 1995;16:428–431. 32 Mannie, MD, Rosser, JM, White, GA: Autologous rat myelin basic protein is a partial agonist that is converted into a full antagonist upon blockade of CD4. Evidence for the integration of efficacious and nonefficacious signals during T cell antigen recognition. J Immunol 1995;154:2642–2654. 33 Renard, V, Delon, J, Luescher, IF, Malissen, B,Vivier, E, Trautmann, A: The CD8 beta polypeptide is required for the recognition of an altered peptide ligand as an agonist. Eur J Immunol 1996;26: 2999–3007. 34 Kersh, GJ, Hedrick, SM: Role of TCR specificity in CD4 versus CD8 lineage commitment. J Immunol 1995;154:1057–1068. 35 Strasser, A, Harris, AW, von Boehmer, H, Cory, S: Positive and negative selection of T cells in T-cell receptor transgenic mice expressing a bcl-2 transgene. Proc Natl Acad Sci USA 1994;91: 1376–1380. 36 Goldrath, AW, Bevan, MJ: Lowaffinity ligands for the TCR drive proliferation of mature CD8+ T
18
37
38 39
40
41
42
43
44
45
46
47
cells in lymphopenic hosts. Immunity 1999;11:183–190. Ernst, B, Lee, DS, Chang, JM, Sprent, J, Surh, CD: The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery. Immunity 1999;11:173–181. Freitas, AA, Rocha, B: Peripheral T cell survival. Curr Opin Immunol 1999;11:152–156. Janeway, CA, Jr.: T-cell development: a role for self-peptides in positive selection. Curr Biol 1999; 9:R342–R345. McGargill, MA, Hogquist, KA: Antigen-induced coreceptor downregulation on thymocytes is not a result of apoptosis. J Immunol 1999;162:1237–1245. Sprent, J, Schaefer, M, Hurd, M, Surh, CD, Ron, Y: Mature murine B and T cells transferred to SCID mice can survive indefinitely and many maintain a virgin phenotype. J Exp Med 1991;174:717–728. Takeda, S, Rodewald, HR, Arakawa, H, Bluethmann, H, Shimizu, T: MHC class II molecules are not required for survival of newly generated CD4+ T cells, but affect their long-term life span. Immunity 1996;5:217–228. Brocker,T: Survival of mature CD4 T lymphocytes is dependent on major histocompatibility complex class II-expressing dendritic cells. J Exp Med 1997;186:1223–1232. Rooke, R, Waltzinger, C, Benoist, C, Mathis, D: Targeted complementation of MHC class II deficiency by intrathymic delivery of recombinant adenoviruses. Immunity 1997;7:123–134. Yamada, H, Ninomiya, T, Hashimoto, A, Tamada, K, Takimoto, H, Nomoto, K: Positive selection of extrathymically developed T cells by self-antigens. J Exp Med 1998; 188:779–784. Tanchot, C, Lemonnier, FA, Perarnau, B, Freitas, AA, Rocha, B: Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 1997;276:2057–2062. Viret, C, Wong, FS, Janeway, CA, Jr.: Designing and maintaining the
48
49
50
51
52
53
54
55 56
57
mature TCR repertoire: the continuum of self-peptide:self-MHC complex recognition. Immunity 1999;10:559–568. Kirberg, J, Berns,A, von Boehmer, H: Peripheral T cell survival requires continual ligation of the T cell receptor to major histocompatibility complex-encoded molecules. J Exp Med 1997;186: 1269–1275. Goodnow, CC: Balancing immunity and tolerance: deleting and tuning lymphocyte repertoires. Proc Natl Acad Sci USA 1996;93: 2264–2271. Rocha, B, von Boehmer, H: Peripheral selection of the T cell repertoire. Science 1991;251: 1225–1228. Ferber, I, Schonrich, G, Schenkel, J, Mellor, AL, Hammerling, GJ, Arnold, B: Levels of peripheral T cell tolerance induced by different doses of tolerogen. Science 1994; 263:674–676. Schonrich, G, Momburg, F, Malissen, M, Schmitt-Verhulst, A-M, Malissen, B, Hammerling, GJ, Arnold, B: Distinct mechansms of extrathymic T cell tolerance due to differential expression of self antigen. Int Immunol 1992;4:581–590. Blish, CA, Dillon, SR, Farr, AG, Fink, PJ:Anergic CD8+ T cells can persist and function in vivo. J Immunol 1999;163:155–164. Dillon, SR, MacKay,VL, Fink, PJ: A functionally compromised intermediate in extrathymic CD8 + T cell deletion. Immunity 1995;3: 321–333. Lenardo, MJ: Fas and the art of lymphocyte maintenance. J Exp Med 1996;183:721–724. Landolfi, MM, Van Houten, N, Russell, JQ, Scollay, R, Parnes, JR, Budd, RC: CD2–CD4–CD8– lymph node T lymphocytes in MRL lpr/ lpr mice are derived from a CD2 +CD4 +CD8 + thymic precursor. J Immunol 1993;151: 1086–1096. Reap, EA, Felix, NJ, Wolthusen, PA, Kotzin, BL, Cohen, PL, Eisenberg, RA: bcl-2 transgenic Lpr mice show profound enhancement of lymphadenopathy. J Immunol 1995;155:5455–5462.
Mannie
58 von Boehmer, H, Kirberg, J, Rocha, B: An unusual lineage of alpha/beta T cells that contains autoreactive cells. J Exp Med 1991; 174:1001–1008. 59 Mehal, WZ, Crispe, IN: TCR ligation on CD8 + T cells creates double-negative cells in vivo. J Immunol 1998;161:1686–1693. 60 Evavold, BD, Allen, PM: Separation of IL-4 production from Th cell proliferation by an altered T cell receptor ligand. Science 1991;252:1308–1310. 61 Evavold, BD, Sloan-Lancaster, J, Hsu, BL, Allen, PM: Separation of T helper 1 clone cytolysis from proliferation and lymphokine production using analog peptides. J Immunol 1993;150:3131–3140. 62 Racioppi, L, Ronchese, F, Matis, LA, Germain, RN: Peptide-major histocompatibility complex class II complexes with mixed agonist/antagonist properties provide evidence for ligand-related differences in T cell receptor-dependent intracellular signaling. J Exp Med 1993;177:1047–1060. 63 Bachmann, MF, Speiser, DE, Zakarian, A, Ohashi, PS: Inhibition of TCR triggering by a spectrum of altered peptide ligands suggests the mechanism for TCR antagonism. Eur J Immunol 1998;28:3110–3119. 64 Sloan-Lancaster, J, Evavold, BD, Allen, PM: Induction of T-cell anergy by altered T-cell-receptor ligand on live antigen-presenting cells. Nature 1993;363:156–159. 65 Sloan-Lancaster, J, Evavold, BD, Allen, PM: Th2 cell clonal anergy as a consequence of partial activation. J Exp Med 1994;180: 1195–1205. 66 Ruiz, PJ, Garren, H, Hirschberg, DL, Langer-Gould,AM, Levite, M, Karpuj, MV, Southwood, S, Sette, A, Conlon, P, Steinman, L: Microbial epitopes act as altered peptide ligands to prevent experimental autoimmune encephalomyelitis. J Exp Med 1999;189:1275–1284. 67 Nicholson, LB, Murtaza,A, Hafler, BP, Sette, A, Kuchroo, VK: A T cell receptor antagonist peptide induces T cells that mediate bystander suppression and prevent
T Cell-Mediated Antigen Presentation and Tolerance
68
69
70
71
72
73
74
75
76
77
autoimmune encephalomyelitis induced with multiple myelin antigens. Proc Natl Acad Sci USA 1997;94:9279–9284. Brocke, S, Gijbels, K, Allegretta, M, Ferber, I, Piercy, C, Blankenstein, T, Martin, R, Utz, U, Karin, N, Mitchell, D:Treatment of experimental encephalomyelitis with a peptide analogue of myelin basic protein. Nature 1996;379: 343–346. Basu, D,Williams, CB,Allen, PM: In vivo antagonism of a T cell response by an endogenously expressed ligand. Proc Natl Acad Sci USA 1998;95:14,332–14,336. Jordan, MS, Riley, MP, von Boehmer, H, Caton, AJ: Anergy and suppression regulate CD4(+) T cell responses to a self peptide. Eur J Immunol 2000;30:136–144. Araneo, BA, Yowell, RL: MHClinked immune response suppression mediated by T cells bearing I-A- encoded determinants. J Immunol 1985;135:73–79. Ben-Nun, A, Strauss, W, Leeman, SA, Cohn, LE, Murre, C, Duby, A, Seidman, JG, Glimcher, LH: An Ia-positive mouse T-cell clone is functional in presenting antigen to other T cells. Immunogenetics 1985;22:123–130. Graf, L, Koch, N, Schirrmacher,V: Expression of Ia antigens in a murine T-lymphoma variant. Mol Immunol 1985;22:1371–1377. Heuer, J, Kolsch, E: Functional studies on the role of I-A molecules expressed by the antigenspecific T suppressor cell clone HF1. J Immunol 1985;134: 4031–4034. Krammer, PH, Hudson, L, Sprent, J: Fc-receptors, Ia-antigens, and immunoglobulin on normal and activated mouse T lymphocytes. J Exp Med 1975;142:1403–1415. Lorber, MI, Loken, MR, Stall,AM, Fitch, FW: I-A antigens on cloned alloreactive murine T lymphocytes are acquired passively. J Immunol 1982;128:2798–2803. Michaelides, M, McKenzie, IF: Detection of Ia alloantigens on phytohemagglutinin-stimulated T blast cells. Transplantation 1981;31:330–333.
78 Osborne, BA, Rudikoff, S: Murine thymocyte and splenocyte Ia antigens are indistinguishable by twodimensional gel electrophoresis. J Immunol 1983;131:1386–1390. 79 Reske-Kunz, AB, Reske, K, Rude, E: Cloned murine Ia + BK-BI2.6.C6 T cells function as accessory cells presenting protein antigens to long-term-cultured antigen-specific T cell lines. J Immunol 1986;136:2033–2040. 80 Sharrow, SO, Mathieson, BJ, Singer,A: Cell surface appearance of unexpected host MHC determinants on thymocytes from radiation bone marrow chimeras. J Immunol 1981;126:1327–1335. 81 Vadas, MA, Miller, JF, McKenzie, IF, Chism, SE, Shen, FW, Boyse, EA, Gamble, JR, Whitelaw, AM: Ly and Ia antigen phenotypes of T cells involved in delayed-type hypersensitivity and in suppression. J Exp Med 1976;144:10–19. 82 Bevan, DJ, Chisholm, PM: Coexpression of CD4 and CD8 molecules and de novo expression of MHC class II antigens on activated rat T cells. Immunol 1986;59: 621–625. 83 Kira, J, Itoyama, Y, Goto, I: Generation of CD4+ blastoid T cells showing marked upregulation of CD4, class I and II MHC, and IL2 receptor molecules is required for the expression of potent encephalitogenicity. Cell Immunol 1989;123:264–275. 84 Reske, K, Mohle, U, Sun, D, Wekerle, H: Synthesis and cell surface display of class II determinants by long-term propagated rat T line cells. Eur J Immunol 1987;17: 909–914. 85 Sopori, ML, Cohen, DA, Cherian, S, Kaplan, AM: Antigen presentation in the rat: role of a nonadherent, nonphagocytic, W3/13, OX-6 positive T cell in the presentation of antigen to primed T lymphocytes. Cell Immunol 1984;87: 177–191. 86 Sopori, ML, Cohen, DA, Cherian, S, Perrone, RS, Kaplan,AM:Antigen presentation in the rat. II. An Ia+ radiosensitive T cell can present antigen to primed Ia– T cells. J Immunol 1985;134:1369–1373.
19
87 Sun, D, Woodland, DL, Coleclough, C, Wendling, U, Reske, K: An MHC class II-expressing T cell clone presenting conventional antigen lacks the ability to present bacterial superantigen. Intern Immunol 1995;7:1079–1085. 88 Reizis, B, Schramm, C, Cohen, IR, Mor, F: Expression of major histocompatibility complex class II molecules in rat T cells. Eur J Immunol 1994;24:2796–2802. 89 Celis, E, Saibara, T: Binding of T cell receptor to major histocompatibility complex class II- peptide complexes at the single-cell level results in the induction of antigen unresponsiveness (anergy). Eur J Immunol 1992;22:3127–3134. 90 Celis, E, Goodwin, JJ, Saibara, T: Peptide-induced proliferation and lymphokine production in human T cells in the absence of antigenpresenting cells: role of T-cell activation state and costimulatory signals. Hum Immunol 1992;34: 173–180. 91 Evans, RL, Faldetta, TJ, Humphreys, RE, Pratt, DM,Yunis, EJ, Schlossman, SF: Peripheral human T cells sensitized in mixed leukocyte culture synthesize and express Ia-like antigens. J Exp Med 1978;148:1440–1445. 92 Grunow, R, Frutig, K, Pichler, WJ: Anergy induction in human CD4+ T-cell clones by stimulation with soluble peptides does not require cell proliferation and is accompanied by elevated IL4 production. Cell Immunol 1996;173:79–86. 93 Ko, HS, Fu, SM, Winchester, RJ, Yu, DT, Kunkel, HG: Ia determinants on stimulated human T lymphocytes. Occurrence on mitogen- and antigen-activated T cells. J Exp Med 1979;150: 246–255. 94 Lamb, JR, Feldmann, M: Essential requirement for major histocompatibility complex recognition in T-cell tolerance induction. Nature 1984;308:72–74. 95 LaSalle, JM, Ota, K, Hafler, DA: Presentation of autoantigen by human T cells. J Immunol 1991; 147:774–780. 96 LaSalle, JM, Toneguzzo, F, Saadeh, M, Golan, DE, Taber, R,
20
97
98
99
100
101
102
103
104 105
Hafler, DA: T-cell presentation of antigen requires cell-to-cell contact for proliferation and anergy induction. Differential MHC requirements for superantigen and autoantigen. J Immunol 1993;151: 649–657. Mauri, D, Wyss-Coray, T, Gallati, H, Pichler, WJ: Antigen-presenting T cells induce the development of cytotoxic CD4+ T cells. I. Involvement of the CD80-CD28 adhesion molecules. J Immunol 1995;155:118–127. Reinherz, EL, Kung, PC, Pesando, JM, Ritz, J, Goldstein, G, Schlossman, SF: Ia determinants on human T-cell subsets defined by monoclonal antibody. Activation stimuli required for expression. J Exp Med 1979;150:1472–1482. Satyaraj, E, Rath, S, Bal, V: Induction of tolerance in freshly isolated alloreactive T cells by activated T cell stimulators is not due to the absence of CD28-B7 interaction. J Immunol 1995;155:4669–4675. Yu, DT, 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;152:89s–98s. LaSalle, JM, Tolentino, PJ, Freeman, GJ, Nadler, LM, Hafler, DA: Early signaling defects in human T cells anergized by T cell presentation of autoantigen. J Exp Med 1992;176:177–186. Satyaraj, E, Rath, S, Bal, V: Induction of tolerance in freshly isolated alloreactive CD4+ T cells by activated T cell stimulators. Eur J Immunol 1994;24:2457–2461. Bettens, F, Frei, E, Frutig, K, Mauri, D, Pichler, WJ, WyssCoray, T: Noncytotoxic human CD4+ T-cell clones presenting and simultaneously responding to an antigen die of apoptosis. Cell Immunol 1995;161:72–78. Pichler,WJ,Wyss-Coray, T: T cells as antigen-presenting cells. Immunol Today 1994;15:312–315. LaSalle, JM, Toneguzzo, F, Saadeh, M, Golan, DE, Taber, R, Hafler, D: T cell presentation of antigen requires cell-to-cell contact for proliferation and anergy induction: differential MHC
106 107
108
109
110
111
112
113
114
requirements for superantigen and autoantigen. J Immunol 1993;151: 649–657. LaSalle, JM, Hafler, DA: T cell anergy. FASEB J 1994;8:601–608. Celis, E, Saibara, T: Binding of T cell receptor to major histocompatibility complex class II peptide complexes at the single-cell level results in the induction of antigen unresponsiveness (anergy). Eur J Immunol 1992;22:3127–3134. Mauri, D, Wyss-Coray, T, Gallati, H, Pichler, WJ: Antigen-presenting T cells induce the development of cytotoxic CD4+ T cells. I. involvement of the CD80–CD28 adhesion molecules. J Immunol 1995;155:118–127. Lamb, JR, Skidmore, BJ, Green, N, Chiller, JM, Feldman, M: Induction of tolerance in influenza virusimmune T lymphocyte clones with synthetic peptides of influenza hemagglutinin. J Exp Med 1983; 157:1434. Mannie, MD, White, GA, Lake, KR, Nardella, JP, Marinakis, CA, McConnell, TJ: T-helper lymphocytes specific for myelin basic protein: low-density activation prolongs a postactivation refractory phase marked by decreased pathogenicity and enhanced sensitivity to anergy. Cell Immunol 1996;172:108–117. Mannie, MD, Rendall, SK,Arnold, PY, Nardella, JP, White, GA: Anergy-associated T cell antigen presentation. A mechanism of infectious tolerance in experimental autoimmune encephalomyelitis. J Immunol 1996;157:1062–1070. Mannie, MD, White, GA, Nardella, JP, Davidian, DK, Arnold, PY: Partial agonism elicits an enduring phase of T-cellmedicated antigen presentation. Cell Immunol 1998;186:83–93. Mannie, MD, Nardella, JP, White, GA, Arnold, PY, Davidian, DK: Class II MHC/peptide complexes on T cell antigen-presenting cells: agonistic antigen recognition inhibits subsequent antigen presentation. Cell Immunol 1998; 186:111–120. Arnold, PY, Davidian, DK, Mannie, MD:Antigen presentation
Mannie
by T cells: T cell receptor ligation promotes antigen acquisition from professional antigen-presenting cells. Eur J Immunol 1997;27: 3198–3205. 115 Walker, MR, White, GA, Nardella, JP, Mannie, MD: An autologous self-antigen differentially regulates expression of I-A glycoproteins and B7 costimulatory molecules on CD4– CD8– T helper cells. J Leukoc Biol 1999;66:120–126. 116 Patel, DM, Arnold, PY, White, GA, Nardella, JP, Mannie, MD:
T Cell-Mediated Antigen Presentation and Tolerance
Class II MHC/peptide complexes are released from APC and are acquired by T cell responders during specific antigen recognition. J Immunol 1999;163:5201–5210. 117 Raposo, G, Nijman, HW, Stoorvogel, W, Liejendekker, R, Harding, CV, Melief, CJ, Geuze, HJ: B lymphocytes secrete antigenpresenting vesicles. J Exp Med 1996;183:1161–1172. 118 Arnold, PY, Mannie, MD:Vesicles bearing MHC class II molecules mediate transfer of antigen from
antigen-presenting cells to CD4+ T cells. Eur J Immunol 1999;29: 1363–1373. 119 Mannie, MD: A unified model for T cell antigen recognition and thymic selection of the T cell repertoire. J Theor Biol 1991;151: 169–192. 120 Mannie, MD: Immunological self/nonself discrimination: integration of self vs nonself during cognate T cell interactions with antigen-presenting cells. Immunol Res 1999;19:65–87.
21