Presentation of exogenous protein antigens on major histocompatibility complex class I molecules by dendritic cells: pathway of presentation and regulation by cytokines

Several recent studies have shown that dendritic cells (DC) pulsed with soluble proteins can present peptide epitopes derived from these exogenous antigens on major histocompatability complex (MHC) class I molecules and induce an antigen-specific cytotoxic T lymphocyte (CTL) response. We provide evidence here that DC use macropinocytosis to capture soluble antigens that are then presented on MHC class I molecules. The presentation of an epitope derived from soluble ovalbumin was transporter associated with antigen presentation (TAP)-dependent, brefeldin A-sensitive, blocked by inhibitors of proteasomes, and resistant to chloroquine. These data suggest that exogenous antigens access the cytosol of DC and are proccessed for presentation via the same pathway described for conventional MHC class I-restricted cytosolic antigens. Proinflammatory mediators such as tumor necrosis factor-alpha (TNF-alpha) and lipopolysaccharide (LPS) reduced the efficiency of ovalbumin presentation via this pathway. This reduced presentation was not due to impaired expression of class I molecules because these substances upregulated the cell surface expression of Kb-molecules comparable to levels induced by interferon-gamma (IFN-gamma) treatment. The addition of IFN-gamma increased ovalbumin presentation even in the presence of TNF-alpha or LPS. These results show that DC might be involved in the cross-priming phenomenon. This could offer the immune system an additional pathway for effective priming of cytotoxic T cells and provide the possibility to activate both CD4 and CD8 T-cell responses.

CTL responses to H2-M3-restricted Listeria epitopes

Cytotoxic T cells (CTL) play a critical role in the murine immune response to Listeria monocytogenes (Listeria). Bacterial antigens are presented to Listeria-specific CTL by products of both conventional, polymorphic MHC class Ia and non-polymorphic MHC class Ib alleles. The H2-M3 class Ib gene product, M3, preferentially presents formylmethionine-initiating (fMet) peptides derived from the N termini of bacterial and mitochondrial proteins. Thus, M3 signals the presence of bacterial invaders to CTL effectors. Listeria-encoded fMet peptide epitopes for H2-M3-restricted CTL have recently been identified. These and other identified fMet peptides are predominantly comprised of hydrophobic residues and appear to be cleaved from membrane-bound proteins. The subcellular location and membrane topology of such proteins may be significant factors in their selection as target antigens for H2-M3-restricted CTL. Such rules may prove useful for prediction of candidate fMet peptide epitopes from other bacterial proteins and species. Studies using synthetic fMet peptides to stimulate CTL ex vivo are also discussed. These latter studies indicate that Listeria infection boosts H2-M3-restricted CTL responses. However, in contrast to MHC class Ia-restricted CTL responses, fMet peptide-specific CTL are observed in a large proportion of cultures from non-immunized, conventionally housed (non-SPF) mice. The CTL activity in these latter cultures may reflect priming in vivo on cross-reactive antigens, or may indicate that requirements for priming of H2-M3-restricted CTL are less stringent than for class Ia-restricted responses.

CD8 lineage commitment in the absence of CD8

The absence of cytotoxic T lymphocyte activity and the failure of MHC class I-restricted T cell receptor (TCR) transgenic thymocytes to mature in CD8alpha-deficient mice suggest that CD8 may be essential for CD8 lineage commitment. We report that variants of the antigenic peptide that delete TCR transgenic thymocytes from CD8 wild-type but not CD8alpha-deficient mice can restore positive selection of CD8 lineage cells in the absence of CD8. The positively selected cells down-regulate CD4, up-regulate TCR, respond to the antigenic peptide, and express CD8beta mRNA. Interestingly, there was no enhanced selection of CD4+ T cells, implying that the TCR-MHC interaction, even in the absence of CD8, provided instructive signaling for commitment to the CD8 lineage. Our results are discussed in terms of recent models of T cell lineage commitment.

Identification of a natural T cell epitope presented by Salmonella-infected macrophages and recognized by T cells from orally immunized mice

Murine infection with Salmonella typhimurium provides models for typhoid fever and long-lasting protective immunity conferred by oral vaccination with viable attenuated bacteria. To further understand the role of T cells in these systems, we identified a bacterial Ag recognized by murine T cells responding to a Salmonella infection. From orally infected mice, we derived a CD4+ Ak-restricted T cell clone (7.4.8) the stimulatory Ag of which was provided by S. typhimurium or its flagella, but not by other salmonellae or S. typhimurium mutants unable to synthesize the flagellar filament protein FliC. We mapped antigenic activity to FliC hypervariable region VI using a generally applicable method of sequential C-terminal truncation of recombinant MalE-FliC fusion proteins. Residues 339-350 are the minimal FliC structure capable of stimulating 7.4.8 and represent the first reported Salmonella-specific epitope recognized by T cells from infected mice. T cells with this specificity are generated by oral immunization, reactivity can be recovered for at least 5 mo afterwards, and FliC is the dominant recall Ag for CD4+ T cells from protectively immunized C3H/HeJ mice. FliC 339-350 is presented by macrophages infected with viable S. typhimurium, and presentation, but not bacterial uptake, is greatly enhanced by pretreatment of macrophages with IFN-gamma. These data point to the importance of IFN-gamma-activated macrophages in the stimulation of T cells responding to facultative intracellular pathogens like S. typhimurium and provide a model system for studying Ag-specific T cell responses in murine salmonellosis.

Identification of a natural T cell epitope presented by Salmonella-infected macrophages and recognized by T cells from orally immunized mice

Murine infection with Salmonella typhimurium provides models for typhoid fever and long-lasting protective immunity conferred by oral vaccination with viable attenuated bacteria. To further understand the role of T cells in these systems, we identified a bacterial Ag recognized by murine T cells responding to a Salmonella infection. From orally infected mice, we derived a CD4+ Ak-restricted T cell clone (7.4.8) the stimulatory Ag of which was provided by S. typhimurium or its flagella, but not by other salmonellae or S. typhimurium mutants unable to synthesize the flagellar filament protein FliC. We mapped antigenic activity to FliC hypervariable region VI using a generally applicable method of sequential C-terminal truncation of recombinant MalE-FliC fusion proteins. Residues 339-350 are the minimal FliC structure capable of stimulating 7.4.8 and represent the first reported Salmonella-specific epitope recognized by T cells from infected mice. T cells with this specificity are generated by oral immunization, reactivity can be recovered for at least 5 mo afterwards, and FliC is the dominant recall Ag for CD4+ T cells from protectively immunized C3H/HeJ mice. FliC 339-350 is presented by macrophages infected with viable S. typhimurium, and presentation, but not bacterial uptake, is greatly enhanced by pretreatment of macrophages with IFN-gamma. These data point to the importance of IFN-gamma-activated macrophages in the stimulation of T cells responding to facultative intracellular pathogens like S. typhimurium and provide a model system for studying Ag-specific T cell responses in murine salmonellosis.

Virus-mediated delivery of antigenic epitopes into dendritic cells as a means to induce CTL

Dendritic cells (DC) are potent inducers of CD8+ T cells and can stimulate protective antitumor immunity when pulsed with an antigenic peptide or protein. We used a replication-deficient adenovirus containing a Kb-restricted antigenic peptide of chicken OVA to study CTL induction in vitro and in vivo after adenovirus-mediated gene transfer into DC. The efficiency of adenovirus-infected DC in eliciting a specific CTL response was compared with immunizations with a recombinant vaccinia virus and DC pulsed with peptide or protein. An immortalized DC line derived from a C57BL/6 mouse and freshly isolated splenic DC from C57BL/6 mice were used in CTL induction. Virus-infected DC elicited the strongest Ag-specific CTL response in vitro and in vivo and induced protective antitumor immunity to a challenge with EG.7 tumors (EL-4 cell line expressing OVA). Direct immunization of mice with recombinant adenovirus resulted in the induction of high titers of neutralizing Abs, which precluded a boost of a CTL response after repeated inoculations. However, repeated injections of virus-infected DC induced only low titers of neutralizing Abs. Furthermore, the presence of neutralizing Abs specific for the virus did not affect the usefulness of infected DC as repeated applications of virus-infected DC boosted the CTL response even in mice previously infected with the recombinant vector. The use of DC infected with a recombinant virus has advantages over other forms of immunization and could provide an alternative approach for designing vaccination therapies.

Virus-mediated delivery of antigenic epitopes into dendritic cells as a means to induce CTL

Dendritic cells (DC) are potent inducers of CD8+ T cells and can stimulate protective antitumor immunity when pulsed with an antigenic peptide or protein. We used a replication-deficient adenovirus containing a Kb-restricted antigenic peptide of chicken OVA to study CTL induction in vitro and in vivo after adenovirus-mediated gene transfer into DC. The efficiency of adenovirus-infected DC in eliciting a specific CTL response was compared with immunizations with a recombinant vaccinia virus and DC pulsed with peptide or protein. An immortalized DC line derived from a C57BL/6 mouse and freshly isolated splenic DC from C57BL/6 mice were used in CTL induction. Virus-infected DC elicited the strongest Ag-specific CTL response in vitro and in vivo and induced protective antitumor immunity to a challenge with EG.7 tumors (EL-4 cell line expressing OVA). Direct immunization of mice with recombinant adenovirus resulted in the induction of high titers of neutralizing Abs, which precluded a boost of a CTL response after repeated inoculations. However, repeated injections of virus-infected DC induced only low titers of neutralizing Abs. Furthermore, the presence of neutralizing Abs specific for the virus did not affect the usefulness of infected DC as repeated applications of virus-infected DC boosted the CTL response even in mice previously infected with the recombinant vector. The use of DC infected with a recombinant virus has advantages over other forms of immunization and could provide an alternative approach for designing vaccination therapies.

CD8 T-cell recognition of macrophages and hepatocytes results in immunity to Listeria monocytogenes

CD8 T cells are effective mediators of specific immunity to infection by Listeria monocytogenes, a bacterial pathogen that initially infects macrophages in the spleen and liver and subsequently spreads to hepatocytes and unidentified parenchymal cells in the spleen. To identify the in vivo target cells of L. monocytogenes-immune CD8 T cells, adoptive transfer assays were performed with bone marrow chimeric or transgenic host mice which had been manipulated to alter the major histocompatibility complex molecules expressed on macrophages or hepatocytes. L. monocytogenes-immune CD8 T cells mediate significant immunity in BDF1-->beta 2 M-/- chimeras, comparable to that seen in unmanipulated BDF1 recipients. L. monocytogenes-immune CD8 T cells also mediate significant antilisterial immunity in parent-->F1 chimeras when the CD8 T cells are syngeneic with the bone marrow donor. These data demonstrate that bone marrow-derived macrophages are major targets for L. monocytogenes-immune CD8 T cells in adoptive transfer assays. Interestingly, significant immunity was observed in parent-->F1 chimeras when the L. monocytogenes-immune CD8 T cells were not syngeneic with the bone marrow donor, suggesting that recognition of Listeria-infected non-bone-marrow-derived cells such as hepatocytes may also occur in vivo. Consistent with this possibility, H-2Kb-restricted CD8 T cells specific for the listeriolysin O molecule mediate significant immunity in the liver, but not the spleen, in transgenic mice expressing H-2Kb only on hepatocytes. In addition, Listeria-specific CD8 T cells lyse Listeria-infected hepatocyte-like cells in vitro. Thus, Listeria-infected hepatocytes can be recognized by CD8 T cells in vivo and in vitro.

Primary and secondary immune responses to Listeria monocytogenes

Recent studies have revealed the complexity of cytokine and cellular interactions required for resistance to primary Listeria monocytogenes infection and have illustrated that resistance to secondary infection may occur through multiple pathways. Analyses of Listeria epitope generation and the specificity of protective CD8(+) T cells have suggested that future research should focus on secreted protein antigens in specific resistance to infection and have increased our understanding of Listeria antigens presented by MHC class l-b molecules.

Identification of an H2-M3-restricted Listeria epitope: implications for antigen presentation by M3

Using expression cloning, we have identified an H2-M3-restricted epitope of the intracellular bacterial pathogen Listeria monocytogenes. Picomolar concentrations of an amino-terminal N-formylated hexapeptide, fMIGWII, targeted cells for lysis by CD8+ cytotoxic T cells, while the nonformylated peptide was approximately 100-fold less active. The sequence of the 185 aa protein source of this epitope predicts a transmembrane protein that retains its N terminus and assumes an N(out)-C(in) topology. This membrane orientation offers an explanation for the protection of the epitope from deformylases present in the bacterial cell and suggests an explanation for the ability of phagocytes to present H2-M3-restricted bacterial epitopes via a vacuolar TAP-independent mechanism.

Positive and negative selection invoke distinct signaling pathways

During T cell development, interaction of the T cell receptor (TCR) with cognate ligands in the thymus may result in either maturation (positive selection) or death (negative selection). The intracellular pathways that control these opposed outcomes are not well characterized. We have generated mice expressing dominant-negative Ras (dnRas) and Mek-1 (dMek) transgenes simultaneously, either in otherwise normal animals, or in animals expressing a transgenic TCR, thereby permitting a comprehensive analysis of peptide-specific selection. In this system, thymocyte maturation beyond the CD4+8+ stage is blocked almost completely, whereas negative selection, assessed using an in vitro deletion protocol, is quantitatively intact. This suggests that activation of the mitogen-activated protein kinase (MAPK) cascade is necessary for positive selection, but irrelevant for negative selection. Generation of gamma/delta and of CD4-8- alpha/beta T cells proceeds normally despite blockade of the MAPK cascade. Hence, only cells that mature via conventional, TCR-mediated repertoire selection require activation of the MAPK pathway to complete their maturation.

Selective activation of Fas/Fas ligand-mediated cytotoxicity by a self peptide

To study how MHC-associated self antigens may regulate the function of T cells in the periphery, we generated CD8+ T cell lines specific for a single residue variant of a self peptide. The self peptide (GAYEFTTL) was isolated from H-2-Kb class I MHC molecules immunopurified from tumor cells. CD8+ CTL lines from H-2b mice were generated against a variant peptide, pE4R, (arginine for glutamic acid at the TCR contact position 4). In short-term 51Cr-release assays, these CTL lysed H-2Kb targets that were pulsed with picomolar levels of pE4R but did not lyse target cells coated with the self peptide at micromolar levels. However, in overnight assays the CTL lysed Fas-positive target cells in the presence of nanomolar levels of the self peptide. This killing was shown to be entirely Fas/Fas ligand mediated by blocking with anti-Fas antibody and Fas-Fc chimeric molecules. While the self peptide was unable to induce serine esterase release from the CTL, it did induce secretion of IFN-gamma. By these criteria then, the unmodified self ligand served as a partial agonist for the CTL raised against a single-residue variant. CD8+ T cell lines raised by in vitro stimulation with the self peptide were likewise unable to kill self peptide-coated targets via the perforin pathway but did lyse targets via Fas. These and similar data from other groups show that self antigens (i.e., MHC/peptide complexes) may be recognized by mature peripheral T cells. The T cell population is tolerant of the self antigen in the sense that they do not respond to physiological levels of the MHC/peptide complex. However, when the level of self antigen is increased (by using synthetic peptide loading) CD8+ T cells may respond by proliferation, IFN-gamma secretion, Fas ligand upregulation, and Fas-mediated cytolysis but are still unable to respond by perforin-mediated cytolysis or granzyme release. The physiological significance of such partial activation in regulation of the immune system remains to be demonstrated.

The nature of the peptide/MHC ligand involved in positive selection

Positive and negative selection in the thymus occur when the TCR on an immature thymocyte engages self peptide/MHC complexes present on epithelial cells. The signaling mechanism which dictates the selection outcome is currently a matter of intense investigation. Here we review experiments that defined the peptide ligands for positive selection. A comparison of these to the peptide ligands that induce negative selection (antigenic peptides) sheds some insight into how a thymocyte interprets peptide/MHC interactions leading to life versus death of the cell.

Increased peptide promiscuity provides a rationale for the lack of N regions in the neonatal T cell repertoire

Making use of mice deficient for terminal deoxynucleotidyl transferase (TdT) expression and a random peptide library, we have examined the diversity and peptide specificity of the neonatal T cell repertoire specific for a single H-2Db-restricted peptide. Consistent with the predicted decrease in repertoire diversity, polyclonal CTL lines and individual clones from different TdTo mice are more similar to each other than those from different wild-type mice in terms of their fingerprints of cross-reactivity to the library and their TCR sequences. We have also found that several TdTo CTL clones cross-react with many more library peptides than wild-type CTL clones. In a few instances, the degree of peptide promiscuity correlates with TCR sequence characteristics such as N region addition and homology-directed recombination, but not CDR3 loop length. Based on epitope titrations for each clone, TCR affinity for antigen is consistently high; thus, this reduced specificity for peptide may coincide with an accentuated affinity for the alpha helices of the MHC. Peptide promiscuity in the neonate may allow the relatively small numbers of T cells in the periphery to protect against a broader range of pathogens.

Murine cytotoxic T lymphocytes induced following Chlamydia trachomatis intraperitoneal or genital tract infection respond to cells infected with multiple serovars

The obligate intracellular bacterium Chlamydia trachomatis is associated with human diseases ranging from blinding trachoma to sexually acquired genital infections and the systemic disease lymphogranuloma venereum (LGV). We have previously reported the isolation and culture of protective murine cytotoxic T lymphocytes (CTL) following intraperitoneal infection with C. trachomatis serovar L2, a serotype associated with human LGV. In this report, we now demonstrate that CTL can also be primed following introduction of C. trachomatis serovar L2 into the uterus or ovarian bursa of mice. We also describe Chlamydia-specific CTL lines isolated following murine infection with a typical human urogenital isolate of C. trachomatis (serovar D) and show that such CTL can be primed by intraperitoneal, intrauterine, or intrabursal infection. Last, we demonstrate that these murine CTL lines respond to multiple serovars, recognizing and lysing cells infected with C. trachomatis serovars B, C, D, F, J, K, L2, and L3, representative of organisms causing blinding trachoma, genital infection, and LGV.

Specific immunity to Listeria monocytogenes in the absence of IFN gamma

Cytokine and cytokine receptor gene knockout mice provide powerful experimental systems to characterize the functions of these molecules in resistance to infectious disease. Such mice may also provide unique models of immune deficiency to learn whether manipulation of the immune response can overcome the specific dysfunction. We demonstrate that resistance of IFN gamma gene knockout (GKO-/-) mice to the intracellular bacterium Listeria monocytogenes is severely impaired compared with wild-type mice. However, immunization of GKO-/- mice with an attenuated L. monocytogenes strain generates antigen-specific CD8 T cell responses that can transfer immunity to naive hosts. Furthermore, vaccinated GKO-/- mice themselves exhibit 20,000-fold increased resistance to challenge with virulent L. monocytogenes and this resistance appears to be CD8 T cell mediated. These studies demonstrate that vaccination-induced immunity can overcome the absence of a cytokine that is critical for resistance to acute infection.

Strong agonist ligands for the T cell receptor do not mediate positive selection of functional CD8+ T cells

Positive selection of functional CD8+ T cells expressing an MHC class I-restricted T cell receptor can be induced in fetal thymus organ culture by class I-binding peptides related to the antigenic peptide ligand. Peptides that act as antagonist or weak agonist/antagonist ligands for mature T cells work efficiently in this regard. In the present study, we have investigated whether low concentrations of the original agonist peptide, or variants that still have a strong agonist activity can also mediate positive selection. The antigenic peptide did not induce positive selection at any concentration tested. A strong agonist variant was capable of stimulating the differentiation of TCRhi CD8+ cells, giving the appearance of phenotypic positive selection. However, these cells lacked biological function, since they could not proliferate in response to antigen. The most efficient positive selection resulted with ligands that did not activate mature T cells or stimulate negative selection.

Positive selection of thymocytes

Differentiation of alpha beta T cell receptor (TCR)-expressing T cells involves an obligatory interaction with self-major histocompatibility complex (MHC) molecules in the thymus. This process, called positive selection, both rescues thymocytes from programmed cell death and induces their differentiation into mature T cells. Another critical event in thymic development is to prevent maturation of hazardous autoreactive T cells; thus, mechanisms exist to eliminate T cells with self-reactive receptors (negative selection). How can these two pathways be distinguished? This question, which has long taxed immunologists, is more opposite because many features of the interactions in positive and negative selection are shared: Both processes are exquisitely MHC-allele specific, they involve MHC-bound peptide recognition, and employ at least some overlapping signal transduction pathways. However, resolution of this paradox has become much more feasible with the advent of powerful systems for withdrawing and reconstituting individual components involved in positive selection. This review describes recent advances in our understanding of the cells, receptors, ligands, and signaling pathways involved in this process. A pivotal part of this puzzle is the basis for discrimination between TCR ligands that induce positive vs negative selection. Recent work suggests that the peptide/MHC ligand for positive selection may bind with low avidity to the TCR. The implications of these data for the nature of T cell recognition during positive selection are discussed below.