Structural basis of Kbm8 alloreactivity. Amino acid substitutions on the beta-pleated floor of the antigen recognition site

We have analyzed the functional significance of the four amino acid differences between the parental H-2Kb and mutant H-2Kbm8 glycoproteins. Six bm8 variants including single substitutions at residues 22, 23, 24, and 30 as well as paired substitutions at residues 23, 30 and 22, 24 were generated and transfected into L cells. Surface expression of these H-2Kb variants was analyzed using monoclonal antibodies which bind to well-defined H-2Kb epitopes. No alterations introduced into the conformational structure of H-2Kb by the amino acid substitutions were detected. The effect of these substitutions on CTL recognition was initially analyzed using the following bulk CTL: either H-2Kb anti-H-2Kbm8, H-2Kbm8 anti-H-2Kb, or third party anti-H-2Kb. The alloreactivity between H-2Kb and H-2Kbm8 is dominated by the amino acid substitution at residue 24 (Glu----Ser). The complete bm8 phenotype, however, also requires the additional substitution at residue 22 (Tyr----Phe). The H-2Kbm8 anti-Kb bulk CTL reacted with both variant H-2Kbm8 molecules containing single substitutions at amino acid positions 22 or 24 but not the variant molecule containing both substitutions. Further analysis using three individual H-2Kbm8 anti-Kb CTL clones indicated the complexity of the self Kbm8 phenotype. Clone 8B1.20 did not react to changes in residues 22 or 24. The 8B1.32 clone reacted with the change at residue 22 but not with the change at residue 24, although the 8B1.54 clone reacted with the change at residue 24 but not with the change at residue 22. The changes in residues 23 (Met----Ile) and/or 30 (Asp----Asn) did not impact significantly on the alloantigenic properties of Kbm8 as determined by both the bulk and cloned CTL populations. According to the three-dimensional class I structure the substitution at amino acid 24 is inaccessible to the TCR. The location of this substitution within the Ag recognition site implies that altered peptide binding, and not a disruption of MHC residues that interact with the TCR, is responsible for the alloreactivity between H-2Kb and H-2Kbm8.     

The functional significance of two amino acid polymorphisms in the antigen-presenting domain of class I MHC molecules. Molecular dissection of Kbm3

The functional properties of two amino acid substitutions, characteristic of the bm3 mutation, in the Kb class I glycoprotein were analyzed in light of the HLA-A2 crystal model. The model predicts that amino acid residues extending into the proposed ligand-binding site or projecting up from the alpha-helices are functional with respect to peptide Ag presentation; whereas those residues pointing away from the site are silent. L cell clones expressing Kb, Kbm3, and derivatives of Kbm3, Kbm3-77 (Asp----Ser "ligand-binding") and Kbm3-89 (Lys----Ala "silent"), were generated for the analysis. Serologic characterization of this panel of cells by using the mAb B8-24-3, EH-144, 20-8-4, K9-136, and Y-25 (Kb but not Kbm3 specific) revealed the loss of the epitopes recognized by these mAb in the Kbm3-89 clone and the retention of these epitopes in the Kbm3-77 clone. Analysis of the L cell clones by using B6 anti-bm3 CTL demonstrated that L cell clones expressing Kbm3 or Kbm3-77 were lysed by these CTL, whereas clones expressing Kb, Kbm3-89, and Ld were not lysed. In reciprocal experiments, bm3 anti-B6 CTL lysed L cell clones expressing Kb or Kbm3-89 but were unable to lyse clones expressing Kbm3, Kbm3-77, and Ld. The results indicate that the substitution at amino acid 89 determines the Kbm3 serologic phenotype, whereas the Kbm3 alloreactive phenotype is primarily determined by the substitution at amino acid 77. These findings are in good agreement with the predictions derived from the x-ray crystal model of the HLA-A2 molecule.     

Introduction of soluble protein into the class I pathway of antigen processing and presentation

In order to investigate how peptides associate with class I major histocompatibility complex (MHC) glycoproteins intracellularly, we generated cytotoxic T lymphocytes (CTL) specific for a readily available soluble protein in association with class I. C57BL/6 (H-2b) mice immunized against a syngeneic tumor cell transfected with chicken ovalbumin (OVA) cDNA gave rise to H-2Kb-restricted CTL specific for the OVA258-276 peptide. This synthetic peptide and CNBr fragments of OVA (242-285 and 242-273) were able to target H-2b cells for lysis by the CTL in a 3 hr assay. Cells incubated with native OVA for up to 24 hr did not become sensitized for recognition and lysis. However, when OVA was introduced directly into the cytoplasm of cells by the osmotic lysis of pinosomes, the Kb restricted determinant formed readily.     

In vivo selection of a lymphocytic choriomeningitis virus variant that affects recognition of the GP33-43 epitope by H-2Db but not H-2Kb

CD8 T cells drive the protective immune response to lymphocytic choriomeningitis virus (LCMV) infection and are thus a determining force in the selection of viral variants. To examine how escape mutations affect the presentation and recognition of overlapping T-cell epitopes, we isolated an LCMV variant that is not recognized by T-cell receptor (TCR)-transgenic H-2Db-restricted LCMV GP33-41-specific cytotoxic T lymphocytes (CTL). The variant virus carried a single-amino-acid substitution (valine to alanine) at position 35 of the viral glycoprotein. This region of the LCMV glycoprotein encodes both the Db-restricted GP33-43 epitope and a second epitope (GP34-42) presented by the Kb molecule. We determined that the V-to-A CTL escape mutant failed to induce a Db GP33-43-specific CTL response and that Db-restricted GP33-43-specific CTL induced by the wild-type LCMV strain were unable to kill target cells infected with the variant LCMV strain. In contrast, the Kb-restricted response was much less affected. We found that the V-to-A substitution severely impaired peptide binding to Db but not to Kb molecules. Strikingly, the V-to-A mutation did not change any of the anchor residues, and the dramatic effect on binding was therefore unexpected. The strong decrease in Db binding explains why the variant virus escapes the Db GP33-43-specific response but still elicits the Kb-restricted response. These findings also illustrate that mutations within regions encoding overlapping T-cell epitopes can differentially affect the presentation and recognition of individual epitopes.     

H-2 Effects on cell-cell interactions in the response to single non-H-2 alloantigens

Mice expressing mutant H-2Kb alleles were tested for their ability to generate cytotoxic effector T-cells specific for the non-H-2 histocompatibility alloantigen H-4.2. Cytotoxic effectors specific for H-4.2 are preferentially restricted by the Kb allele. Mutant Kb alleles were observed to differentially regulate the magnitude of the H-4.2-specific cytotoxic effector response. Mice expressing the Kbm5, Kbm6, Kbm7, and Kbm9 alleles generated cytotoxic T-cells to the same level as mice expressing the wild-type Kb allele. Kbm8 and Kbm11 responders generated intermediate levels of effectors, whereas Kbm1, Kbm3, and Kbm10 responders did not generate detectable levels of cytotoxic effectors. Kbm4 responders produced high levels of H-4.2-specific cytotoxic effectors that were variably reactive with wild-type Kb antigens with no H-4.2. The ability to generate H-4.2-specific effectors generally correlated with (1) the ability of mutant Kb molecules to present H-4.2 to wild-type Kb-restricted effectors, and (2) the position of the respective amino acid interchanges on the Kb molecule. Mutations that altered the amino acid sequence in the vicinity of the disulfide bond in the C1 domain had the greatest deleterious effects on Kb-controlled responsiveness to H-4.2. The only exception was the Kbm11 intermediate responder, which differs from Kbm3 in both responsiveness and in a single amino acid interchange. Therefore, the amino acid sequence in the vicinity of the disulfide bond in the C1 domain plays a prominent role in determining the H-4.2-specific immune response potential. These observations are the first to clearly demonstrate association between particular MHC gene product, amino acid sequences and immune responsiveness.     

Recognition of H-2Kb mutant target cells by Moloney virus-specific cytotoxic T lymphocytes from bm13 (H-2Db mutant) mice. I. Full recognition of Kbm11 by Kb-restricted CTL

In C57BL/6 (B6, H-2b) mice, the secondary in vitro CTL response against Moloney leukemia virus is restricted and regulated by the H-2Db locus. B6.C-H- 2bm13 ( bm13 ) mice, however, carrying a mutation at the Db locus, show an increased H-2Kb-restricted CTL response without a demonstrable CTL component restricted by the mutant Dbm13 molecule (D----K shift). These purely Kb-restricted bm13 virus-specific CTL were incubated with a series of Kb mutant virus-infected target cells to study the effect of the mutations at the target cell level. Of six Kb-mutant virus-infected target cells tested, bm1 cells were not recognized and bm8 cells were recognized only marginally by bm13 virus-specific CTL, whereas bm3 , bm5 , bm6 , and bm11 cells were fully recognized. Thus, the bm3 , bm5 , bm6 , and bm11 Kb mutants fully share the relevant H-2K restriction specificities with H-2Kb, whereas the bm1 mutant totally and the bm8 mutant almost completely lack these specificities. This result differs markedly from the restriction site relationships among B6 and these Kb mutants in other antigenic systems. The most striking example concerns the bm11 mutant, which is fully recognized by Moloney-specific CTL, but not at all by Sendai, minor H (H-3.1, H-4.2), and sulfhydryl hapten-specific CTL. Monoclonal anti-H-2Kb antibody B8-3-24 inhibited virus-specific lysis by bm13 CTL of all Kb virus-infected mutant target cells to which this antibody binds. Lysis of bm5 and bm11 but not of bm3 target cells was inhibited, in line with the fact that B8-3-24 antibody does not bind bm3 . On the other hand, not only bm5 and bm11 but also bm3 virus-infected target cells blocked virus-specific lysis to the same extent as syngeneic bm13 target cells. Therefore, bm13 virus-specific CTL populations do not recognize the discrete cluster alteration in the Kbm3 molecule, as identified by antibody B8-3-24. The bm1 and the bm8 mutations, which have structural alterations in completely different sites of the Kb molecule, show complete or almost complete loss, respectively, of Kb-Moloney restriction sites. This finding supports the notion that these virus-specific CTL recognize conformational determinants rather than linear amino acid sequences.     

Saponin adjuvant induction of ovalbumin-specific CD8+ cytotoxic T lymphocyte responses.

The ability of a saponin adjuvant, QS-21, to induce OVA-specific, class I MHC Ag-restricted CTL was investigated using different forms of soluble OVA and OVA adsorbed onto alum as immunogens. C57BL/6 mice were immunized with soluble native or denatured OVA in formulations that contained increasing quantities of QS-21, and CTL responses were measured using EL4 and E.G7-OVA cells as targets and splenic mononuclear cells as effectors. Ag-specific CTL responses were produced but only if the QS-21 adjuvant was used. Similar responses were induced using alum-adsorbed OVA when mixed with the QS-21 adjuvant but not when used alone. The CTL were specific for an epitope present on the OVA258-276 synthetic peptide, which contains the dominant CTL epitope recognized by C57BL/6 mice. The CD8+ subpopulation of lymphocytes in immune mice was not increased in spleens but increased significantly in vitro after culture with soluble OVA. The CTL activity of splenic mononuclear cell preparations was totally destroyed by treatment with mAb specific to the CD8 Ag plus complement. The ability of the QS-21 adjuvant to induce class I MHC Ag-restricted CTL after immunization with soluble proteins is a characteristic unique to saponin adjuvants.     

H-2Kb mutations limit the CTL response to SV40 TASA

The cytotoxic T lymphocyte (CTL) responses directed towards SV40 tumor-associated specific antigen (TASA) in nine strains of spontaneously arising Kb mutant mice were analyzed. All nine mutants generated normal levels of H-2Db-restricted response, but the K-end-restricted CTL response varied. B6.C-H-2bm1 (bm1) did not produce K-end-restricted SV40 TASA-specific CTL upon immunization, and SV40-transformed bm1 cells were not lysed by intra-H-2 recombinant Kb [B10.A(5R)] CTL. Nonreciprocal cross-reactive lysis was seen between B6-H-2bm8 (bm8) and B10.A(5R). Strain B6-H-2bm8 mice produce highly specific Kbm8-restricted CTL that lyse SV40-transformed bm8 cells (Kbm8SV) but not B10.A(5R) target cells (K5RSV), although Kbm8SV targets can be partially lysed by B10.A(5R) CTL. The other seven Kb mutants cross-react with B10.A(5R). These experiments definitively show that genes mapping to the K and/or D region directly control the H-2-restricted CTL response to SV40 TASA.     

Class I-restricted presentation occurs without internalization or processing of exogenous antigenic peptides

Previous studies have shown that glutaraldehyde-fixed cells can present fragmented, but not native, Ag to class II-restricted T cells. This presumably occurs via direct binding of peptides to class II molecules at the cell surface. More recently, it has been shown that viable target cells can present peptides and endogenous, but not exogenous, protein Ag in association with class I MHC molecules to CTL. We have derived CTL specific for a chicken OVA peptide (OVA258-276) recognized in association with H-2Kb. These CTL recognize target cells that endogenously synthesize OVA and cells "loaded" with native OVA but fail to recognize target cells in the presence of exogenous native OVA. Thus, OVA must be intracellularly located to be processed and presented for CTL recognition. It remains unclear, however, whether exogenous peptides require internalization and further processing by target cells or are able to associate directly with class I molecules at the cell surface for CTL recognition. We provide evidence that glutaraldehyde-fixed cells can present synthetic peptides to H-2Kb- and H-2Db-restricted CTL and that such presentation does not require internalization or processing. The peptides used range in size from 16 to 48 amino acids in length. In contrast, glutaraldehyde-fixed cells are incapable of presenting Ag to CTL specific for influenza nucleoprotein and OVA if the cells are fixed within 1 h of viral influenza infection or loading with OVA. Thus, CTL recognition of antigenic peptides appears to occur via direct binding of peptides to class I molecules at the cell surface and does not require any intracellular processing events.     

Clonal analysis of H-2Kb + TNP recognition by T cells with the use of H-2Kbm mutants and H-2Kb-specific monoclonal antibodies

Eleven long-term cytotoxic T lymphocyte (CTL) clones derived from C57BL/10 T cells sensitized in vivo and in vitro with trinitrobenzene sulfonate- (TNBS) treated syngeneic cells were all restricted to the K end of H-2b. The fine specificity of these CTL clones was analyzed by using H-2Kbm mutant target cells and H-2Kb-specific monoclonal antibodies (mAb). Seven distinct patterns of reactivity of the T cell clones could be observed with the use of six H-2Kbm mutant target cells. Further heterogeneity could be detected in terms of the ability of anti-Lyt-2 mAb to inhibit CTL activity. Cross-reactivity between H-2Kb + TNP and H-2Kbm + TNP was observed for all clones tested for bm5 and bm6, but less frequently for bm3 (8/11), bm8 (7/10), bm4 (4/11), and bm1 (3/11). It was further observed that amino acid substitutions located in the first domain only (one clone), or in the second domain only (six clones), or in either the first or the second domain (three clones) of the H-2Kb molecule could affect target cell recognition by a given T cell clone. the latter type of reactivity suggested that some clones recognized "conformational" determinants of the H-2 molecule, or that amino acid substitutions in one domain might influence the structure of the next domain. One H-2Kb + TNP-reactive clone exhibited a heteroclitic behavior with decreasing avidities for target cells expressing H-2Kbm8 + TNP, H-2Kb + TNP, and H-2Kbm8, which further extends the various patterns of T cell cross-reactions observed within a given class of MHC products. The use of H-2Kb-specific mAb in blocking studies as an attempt to define further the H-2Kb epitopes recognized by CTL clones indicated that: a) TNBS treatment may affect the antigenicity of the H-2Kb molecule as assessed by some mAb; and b) that the T cell clone-target cell interaction may or may not be inhibited by a given mAb, depending on structural variations of the H-2Kb molecule (use of H-2Kbm mutants) that do not affect the interaction itself. These results indicate that this type of analysis does not permit correlation of serologic- and T cell-defined epitopes.     

Calcineurin inhibitors block MHC-restricted antigen presentation in vivo

APCs, like T cells, are affected by calcineurin inhibitors. In this study, we show that calcineurin inhibitors efficiently block MHC-restricted exogenous Ag presentation in vivo. Mice were injected with clinical doses of tacrolimus (FK-506) followed by soluble OVA, and dendritic cells (DCs) were isolated from lymph nodes and spleens. The efficacy of OVA peptide presentation by DCs was evaluated using OVA-specific CD8 and CD4 T cells. Tacrolimus inhibited both class I- and class II-restricted DC presentation of OVA to T cells. Tacrolimus also inhibited both class I- and class II-restricted presentation of OVA in peritoneal macrophages isolated from mice injected with tacrolimus followed by soluble OVA. Tacrolimus-treated peritoneal macrophages, however, were able to present synthetic OVA peptide, SIINFEKL. Inclusion of cyclosporine A to biodegradable microspheres containing OVA greatly reduced their capacity to induce OVA-specific CTL response in mice. These findings provide novel insight into the mode of action of calcineurin inhibitors and have important implications for clinical immunosuppression regimens.     

Characterization of the spontaneous mutant H-2Kbm29 indicates that gene conversion in H-2 occurs at a higher frequency than detected by skin grafting.

A spontaneous mutation of H-2Kb, Kbm29, was discovered among the progeny of F1 hybrid parents. Unlike other characterized spontaneous class I variants, this mutant was detected with the use of antibody, rather than tissue grafting. Although Kbm29 is serologically indistinguishable from the previously described mutant molecule Kbm3, it is identical to the parental Kb by skin grafting and CTL assays. A full length cDNA of Kbm29 was amplified by polymerase chain reaction with locus-specific primers, cloned, and sequenced. Two nucleotides were found to be mutated, resulting in a single amino acid change (Lys----Ala) at amino acid 89 of the mature glycoprotein. This is consistent with the observed serologic changes, as the same amino acid substitution is responsible for the serologic profile of Kbm3. The occurrence of a mutation which is not detectable by the methods normally used to screen for H-2 mutants provides evidence that the high spontaneous rate of structural mutation described for the Kb molecule is underestimated.     

Interleukin-4-mediated downregulation of cytotoxic T lymphocyte activity is associated with reduced proliferation of antigen-specific CD8+ T cells

During virus infection, exogenous IL-4 strongly downregulates expression of antiviral cytokines and cytotoxic T lymphocyte (CTL) responses. In this study, we have employed a T cell receptor (TCR) transgenic system to more closely investigate the effect of IL-4 on CTL activity. This system involves mice transgenic for an H2-Kb restricted TCR recognising an ovalbumin (OVA)-specific peptide (OT-I mice), and recombinant vaccinia viruses expressing the gene for OVA (VV-OVA), or OVA together with IL-4 (VV-OVA-IL-4). Spleen cells from OT-I mice were adoptively transferred to irradiated C57BL/6 mice infected with VV-OVA or VV-OVA-IL-4. Five days following transfer, markedly stronger CTL activity was detected in VV-OVA- than in VV-OVA-IL-4-infected recipients. The reduction in CTL activity was associated with a reduction in the number of OVA-specific CD8+ T cells. Proliferation of cells from VV-OVA-IL-4-infected recipients was dramatically reduced, and this is a likely explanation for the IL-4-mediated reduction in the total number of OVA-specific cells and the reduced cytotoxic activity. On a per cell basis, the production of IFNgamma and cytotoxic activity of OVA-specific CD8+ cells was not influenced by IL-4. Taken together, our results indicate that the reduction in CTL activity by exogenous IL-4 is due to a reduced number of antigen-specific effectors, and does not involve a downregulation of effector function of these cells.     

Thermolabile H-2Kb molecules expressed by transporter associated with antigen processing-deficient RMA-S cells are occupied by low-affinity peptides.

RMA-S cells do not express functional TAP, yet they express MHC class I molecules at the cell surface, especially at reduced temperatures (26 degrees C). It is generally assumed that such class I molecules are "empty," devoid of any associated peptide. A radiochemical approach was used to label class I-associated peptides and to determine the extent to which Kb molecules in RMA-S cells are associated with peptides. These studies revealed that at 26 degrees C Kb molecules in RMA-S cells are occupied with self-peptides. Such peptides stably associate with Kb at 26 degrees C but easily dissociate from them at 37 degrees C, suggesting low-affinity interactions between Kb and the associated peptides. At 26 degrees C, at least some of these Kb molecules are stably expressed in a peptide-receptive state on the cell surface, whereas at 37 degrees C they are short lived and are only transiently capable of binding and presenting exogenously supplied OVA 257-264 peptide for presentation to CD8+ Kb-restricted T lymphocytes. Thus contrary to current models of class I assembly in TAP-deficient RMA-S cells, the presumably "empty" molecules are in fact associated with peptides at 26 degrees C. Together, our data support the existence of an alternative mechanism of peptide binding and display by MHC class I molecules in TAP-deficient cells that could explain their ability to present Ag.     

Characterization of the ovalbumin-specific TCR transgenic line OT-I: MHC elements for positive and negative selection

The present report provides the first extensive characterization of the OT-I TCR transgenic line, which produces MHC class I-restricted, ovalbumin-specific, CD8+ T cells (OT-I cells). These cells are shown to be positively selected in vivo in H-2b C57BL/6 mice and in bm5 mice, which express the Kbm5 mutant molecule. In contrast, OT-I cells were not selected by mutant Kb molecules in bm1, bm3, bm8, bm10, bm11 or bm23 mice. Interestingly, however, when positive selection was examined in vitro in foetal thymic organ culture (FTOC), bm1 and bm8 were still poorly selective, but the bm3 haplotype now selected as efficiently as B6. The ability to select in vitro correlated with the capacity to present the ovalbumin (OVA) peptide to OT-I cells, as measured by induction of an OVA-specific proliferative response. These results suggest that a lower affinity TCR:MHC interaction may be necessary for positive selection in FTOC compared with selection in situ.     

Positive selection is limited by available peptide-dependent MHC conformations

Recent data suggest that the diversity of self peptides presented in the thymus during development contributes to positive selection of a diverse T cell repertoire. We sought to determine whether a previously defined "hole in the immunological repertoire" could be explained by the absence of an appropriate selecting self peptide. The repertoire defect in question is the inability of bm8 mice to make an H-2K-restricted response to OVA. Like other OVA-specific, H-2K-restricted receptors, OT-I-transgenic T cells are not positively selected in bm8 mice. Using criteria we had previously established for identifying positive selection ligands, we found peptides that could restore positive selection of OT-I thymocytes in bm8 mice. Thus, the T cell repertoire can be limited by a requirement for specific self peptides during development. Data with MHC-specific Abs suggested that peptides might be able to force MHC residues to adopt different conformations in Kb vs Kbm8. This shows that peptides can potentially contribute to ligand diversity both directly (via variability in the solvent-exposed side chains) and indirectly (through their effect on the MHC conformation). Our data support a model where self peptide diversity allows selection of T cells specific for a broad range of MHC conformations.     

Archaeosomes induce long-term CD8+ cytotoxic T cell response to entrapped soluble protein by the exogenous cytosolic pathway, in the absence of CD4+ T cell help

The unique ether glycerolipids of ARCHAEA: can be formulated into vesicles (archaeosomes) with strong adjuvant activity for MHC class II presentation. Herein, we assess the ability of archaeosomes to facilitate MHC class I presentation of entrapped protein Ag. Immunization of mice with OVA entrapped in archaeosomes resulted in a potent Ag-specific CD8(+) T cell response, as measured by IFN-gamma production and cytolytic activity toward the immunodominant CTL epitope OVA(257-264). In contrast, administration of OVA with aluminum hydroxide or entrapped in conventional ester-phospholipid liposomes failed to evoke significant CTL response. The archaeosome-mediated CD8(+) T cell response was primarily perforin dependent because CTL activity was undetectable in perforin-deficient mice. Interestingly, a long-term CTL response was generated with a low Ag dose even in CD4(+) T cell deficient mice, indicating that the archaeosomes could mediate a potent T helper cell-independent CD8(+) T cell response. Macrophages incubated in vitro with OVA archaeosomes strongly stimulated cytokine production by OVA-specific CD8(+) T cells, indicating that archaeosomes efficiently delivered entrapped protein for MHC class I presentation. This processing of Ag was Brefeldin A sensitive, suggesting that the peptides were transported through the endoplasmic reticulum and presented by the cytosolic MHC class I pathway. Finally, archaeosomes induced a potent memory CTL response to OVA even 154 days after immunization. This correlated to strong Ag-specific up-regulation of CD44 on splenic CD8(+) T cells. Thus, delivery of proteins in self-adjuvanting archaeosomes represents a novel strategy for targeting exogenous Ags to the MHC class I pathway for induction of CTL response.     

Affinity thresholds for naive CD8+ CTL activation by peptides and engineered influenza A viruses

High-avidity interactions between TCRs and peptide + class I MHC (pMHCI) epitopes drive CTL activation and expansion. Intriguing questions remain concerning the constraints determining optimal TCR/pMHCI binding. The present analysis uses the TCR transgenic OT-I model to assess how varying profiles of TCR/pMHCI avidity influence naive CTL proliferation and the acquisition of effector function following exposure to the cognate H-2K(b)/OVA(257-264) (SIINFEKL) epitope and to mutants provided as peptide or in engineered influenza A viruses. Stimulating naive OT-I CD8(+) T cells in vitro with SIINFEKL induced full CTL proliferation and differentiation that was largely independent of any need for costimulation. By contrast, in vitro activation with the low-affinity EIINFEKL or SIIGFEKL ligands depended on the provision of IL-2 and other costimulatory signals. Importantly, although they did generate potent endogenous responses, infection of mice with influenza A viruses expressing these same OVA(257) variants failed to induce the activation of adoptively transferred naive OT-I CTLps, an effect that was only partially overcome by priming with a lipopeptide vaccine. Subsequent structural and biophysical analysis of H2-K(b)OVA(257), H2-K(b)E1, and H2-K(b)G4 established that these variations introduce small changes at the pMHCI interface and decrease epitope stability in ways that would likely impact cell surface presentation and recognition. Overall, it seems that there is an activation threshold for naive CTLps, that minimal alterations in peptide sequence can have profound effects, and that the antigenic requirements for the in vitro and in vivo induction of CTL proliferation and effector function differ substantially.     

MHC class I-restricted cytotoxic lymphocyte responses induced by enterotoxin-based mucosal adjuvants

The ability of enterotoxin-based mucosal adjuvants to induce CD8+ MHC class I-restricted CTL responses to a codelivered bystander Ag was examined. Escherichia coli heat-labile toxin (LT), or derivatives of LT carrying mutations in the A subunit (LTR72, LTK63), were tested in parallel with cholera toxin (CT) or a fusion protein consisting of the A1 subunit of CT fused to the Ig binding domain of Staphylococcus aureus protein A (called CTA1-DD). Intranasal (i.n.) immunization of C57BL/6 mice with CT, CTA1-DD, LT, LTR72, LTK63, but not rLT-B, elicited MHC class I-restricted CD8+ T cell responses to coadministered OVA or the OVA CTL peptide SIINFEKL (OVA257-264). CT, LT, and LTR72 also induced CTL responses to OVA after s.c. or oral coimmunization whereas LTK63 only activated responses after s.c. coimmunization. rLT-B was unable to adjuvant CTL responses to OVA or OVA257-264 administered by any route. Mice treated with an anti-CD4 mAb to deplete CD4+ T cells mounted significant OVA-specific CTL responses after i.n. coadministration of LT with OVA or OVA257-264. Both 51Cr release assays and IFN-gamma enzyme-linked immunospot assays indicated that IFN-gamma-/- and IL-12 p40-/- gene knockout mice developed CTL responses equivalent to those detected in normal C57BL/6 mice. The results highlight the versatility of toxin-based adjuvants and suggest that LT potentiates CTL responses independently of IL-12 and IFN-gamma and probably by a mechanism unrelated to cross-priming.     

Murine B16 melanomas expressing high levels of the chemokine stromal-derived factor-1/CXCL12 induce tumor-specific T cell chemorepulsion and escape from immune control

The chemokine, stromal-derived factor-1/CXCL12, is expressed by normal and neoplastic tissues and is involved in tumor growth, metastasis, and modulation of tumor immunity. T cell-mediated tumor immunity depends on the migration and colocalization of CTL with tumor cells, a process regulated by chemokines and adhesion molecules. It has been demonstrated that T cells are repelled by high concentrations of the chemokine CXCL12 via a concentration-dependent and CXCR4 receptor-mediated mechanism, termed chemorepulsion or fugetaxis. We proposed that repulsion of tumor Ag-specific T cells from a tumor expressing high levels of CXCL12 allows the tumor to evade immune control. Murine B16/OVA melanoma cells (H2b) were engineered to constitutively express CXCL12. Immunization of C57BL/6 mice with B16/OVA cells lead to destruction of B16/OVA tumors expressing no or low levels of CXCL12 but not tumors expressing high levels of the chemokine. Early recruitment of adoptively transferred OVA-specific CTL into B16/OVA tumors expressing high levels of CXCL12 was significantly reduced in comparison to B16/OVA tumors, and this reduction was reversed when tumor-specific CTLs were pretreated with the specific CXCR4 antagonist, AMD3100. Memory OVA-specific CD8+ T cells demonstrated antitumor activity against B16/OVA tumors but not B16/OVA.CXCL12-high tumors. Expression of high levels of CXCL12 by B16/OVA cells significantly reduced CTL colocalization with and killing of target cells in vitro in a CXCR4-dependent manner. The repulsion of tumor Ag-specific T cells away from melanomas expressing CXCL12 confirms the chemorepellent activity of high concentrations of CXCL12 and may represent a novel mechanism by which certain tumors evade the immune system.