Low-avidity self-specific T cells display a pronounced expansion defect that can be overcome by altered peptide ligands.

Thymic expression of self-Ags results in the deletion of high-avidity self-specific T cells, but, at least for certain Ags, a residual population of self-specific T cells with low-affinity TCRs remains after negative selection. Such self-specific T cells are thought to play a role in the induction of T cell-mediated autoimmunity, but may also be used for the induction of antitumor immunity against self-Ags. In this study, we examine the functional competence of a polyclonal population of self-specific CD8+ T cells. We show that low-affinity interactions between TCR and peptide are associated with selective loss of critical T cell functions. Triggering of low levels of IFN-gamma production and cytolytic activity through low-affinity TCRs readily occurs provided high Ag doses are used, but IL-2 production and clonal expansion are severely reduced at all Ag doses. Remarkably, a single peptide variant can form an improved ligand for the highly diverse population of low-avidity self-specific T cells and can improve their proliferative capacity. These data provide insight into the inherent limitations of self-specific T cell responses through low-avidity TCR signals and the effect of modified peptide ligands on self-specific T cell immunity.

A redundant role of the CD3 gamma-immunoreceptor tyrosine-based activation motif in mature T cell function.

At least four different CD3 polypeptide chains are contained within the mature TCR complex, each encompassing one (CD3gamma, CD3delta, and CD3epsilon) or three (CD3zeta) immunoreceptor tyrosine-based activation motifs (ITAMs) within their cytoplasmic domains. Why so many ITAMs are required is unresolved: it has been speculated that the different ITAMs function in signal specification, but they may also serve in signal amplification. Because the CD3zeta chains do not contribute unique signaling functions to the TCR, and because the ITAMs of the CD3-gammadeltaepsilon module alone can endow the TCR with normal signaling capacity, it thus becomes important to examine how the CD3gamma-, delta-, and epsilon-ITAMs regulate TCR signaling. We here report on the role of the CD3gamma chain and the CD3gamma-ITAM in peripheral T cell activation and differentiation to effector function. All T cell responses were reduced or abrogated in T cells derived from CD3gamma null-mutant mice, probably because of decreased expression levels of the mature TCR complex lacking CD3gamma. Consistent with this explanation, T cell responses proceed undisturbed in the absence of a functional CD3gamma-ITAM. Loss of integrity of the CD3gamma-ITAM only slightly impaired the regulation of expression of activation markers, suggesting a quantitative contribution of the CD3gamma-ITAM in this process. Nevertheless, the induction of an in vivo T cell response in influenza A virus-infected CD3gamma-ITAM-deficient mice proceeds normally. Therefore, if ITAMs can function in signal specification, it is likely that either the CD3delta and/or the CD3epsilon chains endow the TCR with qualitatively unique signaling functions.

Circumventing T-cell tolerance to tumour antigens.

During past decades, many attempts have been made to induce or enhance tumour-specific T-cell immunity in cancer patients by vaccination. However, it has become apparent that in a large number of cases the naturally occurring tumour-specific T-cell repertoire is of low affinity and therefore inefficient in mediating tumour rejection. Because of the potential therapeutic value of high affinity TCRs with tumour/lineage specificities, we set out to develop a number of new technologies that can be used to create improved tumour-specific T-cell immunity. These strategies entail: (i) the efficient expansion of low affinity T cells specific for self antigens through the use of variant peptides with improved TCR-binding characteristics; (ii) a retroviral library-based technology to improve the affinity of (self-specific) T-cell receptors in vitro, and (iii) proof of principle for the feasibility of TCR gene transfer as a means to generate T-cell populations with a desired antigen-specificity in vivo. Collectively this toolbox should allow us to create improved T-cell receptors for human tumour antigens, which can subsequently be used to impose tumour-reactivity on to peripheral T cells.

Tumor size at the time of adoptive transfer determines whether tumor rejection occurs.

Here we investigate the minimal requirements for induction of an anti-tumor response in CD8 T cells in vivo. We compare the efficacy of adoptive transfer of CD8 T cells with a transgenic TCR specific for the main cytotoxic T lymphocyte epitope of the influenza virus nucleoprotein (NP) on the growth of NP-expressing EL4 tumors under different conditions. In a setting in which tumor rejection is solely dependent on tumor-specific CD8 T cells, small immunogenic tumors fail to induce a rejection response, despite the fact that they are not ignored: tumor-specific CD8 T cells are activated, differentiate into effector cells and infiltrate the tumor bed. Nevertheless, tumor rejection does not occur. In sharp contrast, the same immunogenic tumor, when growing as a large tumor mass, is rejected by transferred tumor-specific CD8 T cells. The main features which distinguish the rejection response to a large tumor mass from the response to a small tumor is that, in the latter case, activated CD8 T cells appear much later, and in much smaller numbers. Efficacy of adoptive transfer is thus dictated by the size of the tumor mass at the time of transfer. These findings predict that treatment of minimal residual disease with adoptive transfer will fail, unless vaccination is also provided at the time of transfer.

Tracing and characterization of the low-avidity self-specific T cell repertoire.

It is well established that expression of self antigens results in the deletion of the functional high-avidity self-specific T cell repertoire. Due to the low frequency of naturally occurring low-avidity self-specific T cells, a detailed evaluation of their ability to survive and differentiate into effector and memory populations in vivo has yet to be obtained. We here employ tetramer technology to characterize and determine the in vivo fate of a self-specific CD8(+) T cell population specific for a ubiquitously expressed T cell epitope. We find that in influenza nucleoprotein (NP)-transgenic mice (B10NP mice) an oligoclonal population of NP(366 - 374)-specific T cells can be triggered by live influenza virus exposure. The main hallmark of this self-specific T cell population is its diminished avidity for the tetrameric MHC / NP peptide complex. These low-avidity T cells are not deleted and do not down-regulate their antigen or CD8 receptors, and exhibit cytolytic activity towards tumor cells expressing NP endogenously. Strikingly, a secondary influenza infection generates a typical memory response in the low-avidity repertoire. The observation that low-avidity T cells persist in vivo and can differentiate into memory T cells underscores their potential role in anti-tumor immunity.

Selective expansion of cross-reactive CD8(+) memory T cells by viral variants.

The role of memory T cells during the immune response against random antigenic variants has not been resolved. Here, we show by simultaneous staining with two tetrameric major histocompatibility complex (MHC)-peptide molecules, that the polyclonal CD8(+) T cell response against a series of natural variants of the influenza A nucleoprotein epitope is completely dominated by infrequent cross-reactive T cells that expand from an original memory population. Based on both biochemical and functional criteria, these cross-reactive cytotoxic T cells productively recognize both the parental and the mutant epitope in vitro and in vivo. These results provide direct evidence that the repertoire of antigen-specific T cells used during an infection critically depends on prior antigen encounters, and indicate that polyclonal memory T cell populations can provide protection against a range of antigenic variants.

Pre-TCR signaling and inactivation of p53 induces crucial cell survival pathways in pre-T cells.

Signaling through the pre-TCR is essential for early T cell development and is severely impaired in mice lacking the CD3 gamma chain of the pre-TCR. We here address the molecular mechanisms underlying this defect. Impaired pre-TCR signaling is shown to be associated with a profound increase in the number of apoptotic CD4- CD8- (DN) thymocytes. Introduction of p53 deficiency into CD3 gamma-deficient mice completely reverses the cell survival defect in CD3 gamma-deficient DN thymocytes and rescues the block in pre-T cell differentiation. In addition, the CD4+ CD8+ (DP) compartment is expanded to its normal size. These findings suggest that the pre-TCR regulates progression through the DNA-damage checkpoint of the DN to DP transition by inactivating p53.

Systemic T cell expansion during localized viral infection.

In a local immune response, the priming and expansion of the antigen-specific T cell population has been thought to largely take place in the draining lymphoid tissue. This model was primarily based on indirect enumeration of antigen-specific T cells by limiting dilution analyses. Here, tetrameric MHC class I complexes were used to evaluate the contribution of different secondary lymphoid organs in a local immune response by following the CD8+ T cell responses against the immunodominant epitopes of influenza A virus and herpes simplex virus-1. Mice were either intranasally infected with influenza A virus and developed pneumonia or were intradermally injected with herpes simplex virus-1. Remarkably, even though these viruses cause a local infection, the spleen of infected animals contains approximately 50-fold more antigen-specific cytotoxic T cells than the draining lymph nodes. Although antigen-specific T cells in spleen appear not to have experienced any recent encounter with antigen, this population is actively dividing, and over time, the formation of a memory T cell population is observed. These data reveal that there is a remarkably large and distinct population of antigen-specific T cells in spleen in the course of a local antigenic challenge. This T cell compartment may not only form the foundation of a memory T cell pool but could also provide a safeguard against systemic spreading of an infection.

Functional analysis of Ran/TC4 as a protein regulating T-cell costimulation.

Antigen (Ag)-triggered activation of T cells requires engagement of both the T-cell Ag receptor and a costimulatory receptor, for which CD28 can function as a prototypical example. CD80 and CD86 represent ligands for this receptor, and although they are present on professional Ag-presenting cells, these molecules are absent from most tumors. Yet some tumors are still able to costimulate a T-cell response, while others cannot. Therefore, a key question concerns the molecular basis for the costimulation of T cells by those tumor cells not expressing the CD28 ligands CD80 and CD86. Upon screening a cDNA library of such a tumor cell line in a transient COS cell transfection assay for costimulatory activity, we identified Ran/TC4 as a protein whose overexpression results in costimulatory activity. Ran/TC4 is a ubiquitously expressed member of the Ras gene superfamily of small guanosine triphosphate-binding proteins and is involved in nuclear transport; Ran/TC4 cDNA-transfected COS cells specifically costimulate CD8 T cells and not CD4 T cells. Transfection of Ran/TC4 into the costimulation-deficient murine RMA lymphoma cell line introduced costimulatory capacity for CD8 T cells and resulted in markedly elevated levels of nuclear Ran/TC4 protein expression. In addition, in vivo priming of mice with Ran/TC4-transfected RMA cells induced protection against wild-type (wt) RMA tumor cells. Ran/TC4-transfected RMA cells and wt RMA tumor cells exhibit comparable in vivo growth rates in mice lacking T and B cells, and Ran/TC4-mediated tumor rejection thus involves B and/or T cells. This possibility is substantiated by the observation that T cells from normal mice challenged with Ran/TC4-transfected RMA cells can mount a cytotoxic T-cell response not only against the Ran/TC4-transfected tumor cells but also against wt RMA tumor cells. Based on these results, we conclude that gene transfer-mediated elevations in Ran/TC4 can confer costimulatory function for CD8 T cells to tumor cells. This finding suggests a novel application of Ran/TC4 as a protein capable of regulating costimulation in tumor cells.

The TNF receptor family member CD27 signals to Jun N-terminal kinase via Traf-2.

CD27 is a lymphocyte-specific member of the TNF receptor (TNFR) family. It is a costimulatory molecule for peripheral T cells, as defined by its ability to enhance the TCR-induced proliferative response. We show here that CD27 augments TCR-induced Jun N-terminal kinase (JNK) activity in primary murine lymph node T cells. To investigate how CD27 couples to JNK, we performed a yeast two hybrid screen with the CD27 cytoplasmic tail. This revealed that CD27 directly associates with Traf-2. Transfection experiments using dominant negative Traf-2 indicated that CD27 communicates with JNK via Traf-2. These findings group CD27 together with other members of the TNFR family, TNFR-1, -2, CD30 and CD40, which have all been shown to couple to Traf proteins. Since Traf proteins have been reported to initiate an anti-apoptotic signaling pathway, our data suggest that CD27 not only regulates proliferation, but also survival of T lymphocytes.

Non-responsiveness of antigen-experienced CD4 T cells reflects more stringent co-stimulatory requirements.

We recently reported that previously activated T cells, irrespective of the nature of the first stimulus they encountered, are unable to respond to Staphylococcal enterotoxin B (SEB), nor to soluble anti-CD3 monoclonal antibody (mAb) presented by splenic antigen-presenting cells (APC). Such previously activated T cells are, however, fully capable of responding to plate-bound anti-CD3 plus splenic APC. These data suggest differential integration of the T-cell receptor (TCR) and co-stimulatory signalling pathways in naive versus antigen-experienced T cells. Consistent with this hypothesis, anti-CD28 mAb restores the proliferative capacity of resting ex vivo CD45RBlo CD4+ T cells (representing previously activated T cells) to both soluble anti-CD3 mAb and SEB. Interestingly, mAb-mediated engagement of cytotoxic T-lymphocyte antigen-4 (CTLA-4) completely negates the rescue effects mediated by anti-CD28 mAb in CD45RBlo cells. Nevertheless, the non-responsiveness of CD45RBlo CD4+ T cells cannot be reversed by anti-CTLA-4 Fab fragments, indicating that it is not related to negative regulatory effects of CTLA-4 engagement itself. Interestingly, the addition of interleukin-2 (IL-2) restores the proliferative capacity of CD45RBlo CD4+ T cells to SEB and soluble anti-CD3 mAb. Moreover, when rescued by IL-2, the cells are less susceptible to the negative regulatory effects of CTLA-4 engagement. Together, these findings suggest that the non-responsiveness of CD45RBlo CD4+ T cells to certain stimuli may be related to inadequate TCR signalling, primarily affecting IL-2 production.

CD40 and CD70 co-stimulate a potent in vivo antitumor T cell response.

In several studies, CD80, a potent co-stimulatory molecule, has been reported to be responsible for the induction of CD8+ antitumor T cell responses by CD80-transfected tumor cells. However, expression of CD80 by tumors not always ensures generation of a T cell-mediated antitumor response. Variables such as the inherent immunogenicity of a tumor and its major histocompatibility complex (MHC) expression status affect the efficacy of this approach. Therefore, in this study two other co-stimulatory ligands, CD40 and CD70, have been investigated for their ability to co-stimulate antitumor responses. The efficacy of CD40 and CD70 is compared with that of CD80, with respect to CD4 and CD8 T cell co-stimulatory capacity in vitro and their ability to induce in vivo antitumor responses. Furthermore, CD40 and CD70 are tested for their capacity to induce a long-lived memory response in vivo, as defined both by induction of tumor-specific cytotoxic T lymphocytes (CTLs) and rejection of wild-type tumor cells. It was found that, despite the fact that CD40 predominantly stimulates CD4 T cells, CD40-transfected MHC class II-negative P815 tumor cells become highly immunogenic and induce long-lasting memory tumor-specific CTLs in vivo. Furthermore, CD40 and CD70 emerge as powerful and even superior alternatives to CD80 for improving tumor immunogenicity in vivo. While the mechanisms by which they do so remain to be defined, these findings suggest additional strategies for immunotherapy.

Cytotoxic T lymphocyte antigen 4 (CTLA-4) interferes with extracellular signal-regulated kinase (ERK) and Jun NH2-terminal kinase (JNK) activation, but does not affect phosphorylation of T cell receptor zeta and ZAP70.

Cytotoxic T lymphocyte antigen 4 (CTLA-4) is an important regulator of T cell homeostasis. Ligation of this receptor leads to prominent downregulation of T cell proliferation, mainly as a consequence of interference with IL-2 production. We here report that CTLA-4 engagement strikingly selectively shuts off activation of downstream T cell receptor (TCR)/CD28 signaling events, i.e., activation of the microtubule-associated protein kinase (MAPKs) ERK and JNK. In sharp contrast, proximal TCR signaling events such as ZAP70 and TCR-zeta chain phosphorylation are not affected by CTLA-4 engagement on activated T cells. Since activation of the ERK and JNK kinases is required for stimulation of interleukin (IL)-2 transcription, these data provide a molecular explanation for the block in IL-2 production imposed by CTLA-4.

Induction of TCR gene rearrangements in uncommitted stem cells by a subset of IL-7 producing, MHC class-II-expressing thymic stromal cells.

The embryonic thymic microenvironment provides the necessary elements for T cell lineage commitment, but the precise role of individual stromal cell components remains to be determined. Here we address the question of which stromal cell types are required for initiation of V-DJ rearrangements of the TCR-beta and TCR-delta locus in CD117+CD45+ uncommitted fetal liver progenitors. We show that fetal thymic stroma alone is necessary and sufficient for induction of TCR-beta and TCR-delta rearrangements. Furthermore, the ability to induce this T cell commitment step is confined to a subset of MHC class II-positive epithelial cells. Thymic stroma derived from mice with a targeted deletion in the IL-7 gene, however, lacks this ability. These findings set the stage for a further definition of the nature of the thymic stromal cell support in the regulation of T cell commitment.

Nonresponsiveness and susceptibility to CTLA-4 of antigen-exposed CD4 T cells are not regulated by the Bcl-2 family of apoptotic mediators, but can be restored by IL-2.

Memory T cells are considered to be less dependent on costimulation and to respond more vigorously to TCR triggering compared to their naive counterparts. We and others, however, observed that memory CD4 T cells display nonresponsiveness to a variety of stimuli, including superantigens and soluble anti-CD3. We now report that CD28-derived costimulation can revert the nonresponsive state of antigen-exposed CD4 T cells. Interestingly, the rescuing effect of CD28 can be completely negated by CTLA-4 engagement. The malfunction of memory T cells is related to increased cell death; the viability can be restored by CD28 engagement and is negatively regulated by CTLA-4 engagement. Importantly, it has been reported that antigen-exposed T cells express lower levels of the anti-apoptotic mediator Bcl-2. In addition, CD28 costimulation was reported to upregulate the expression levels of Bcl-xL. We therefore examined the possible role of Bcl-2 family proteins in the nonresponsiveness of antigen-exposed CD4 T cells, and determined whether CTLA-4, in analogy to CD28, mediates its negative regulatory effects via the Bcl-2 family of apoptotic mediators. Our data indicate that neither the nonresponsiveness nor the susceptibility to CTLA-4 of antigen-experienced CD4 T cells are related to the expression levels of Bcl-2 or Bax. The rescuing effects of CD28, however, may be related to increased Bcl-xL levels. Addition of IL-2 normalizes the nonresponsiveness of memory CD4 T cells and renders these cells resistant to the negative effects of CTLA-4 engagement. Impaired IL-2 production is therefore likely to be the cause of the malfunction and CTLA-4 susceptibility of memory CD4 T cells.

CD28-B7 interactions function to co-stimulate clonal deletion of double-positive thymocytes.

Negative selection of thymocytes only occurs if next to signals through the TCR, additional antigen-presenting cell (APC)-derived signals are also provided. It has been unclear which molecular interactions lead to the generation of these signals. In particular, the involvement of CD28 and its ligands B7-1 and B7-2 has been controversial. In the present study, we re-address this issue and first confirm that cross-linking CD28 molecules on thymocytes can indeed complement TCR-derived signals for induction of deletion upon TCR engagement with antibodies. Furthermore, we extend these findings by documenting that also peptide agonist-induced deletion can be co-stimulated by antibody-mediated engagement of CD28. Additionally, blocking B7-1 or B7-2 reduces negative selection induced by both anti-CD3 and peptide agonist in suspension cultures and in fetal thymic organ culture. At the same time, prominent co-stimulation of TCR-induced deletion could be provided by a B7-negative cell line. Together these results definitively demonstrate that CD28-B7 interactions can function to co-stimulate induction of clonal deletion, while yet to be identified B7-independent co-stimulatory signals can fulfil this function as well.

Expression of tyrosine kinase gene in mouse thymic stromal cells.

Amongst the most important signal transduction molecules involved in regulating growth and differentiation are the protein tyrosine kinases (PTK). Since T cell development is a consequence of interactions between thymic stromal cells (TSC) and thymocytes, identification of the PTK in both compartments is required to dissect the mechanisms that control this process. Here we report a search for PTK in mouse TSC, using RT-PCR to survey the repertoire of PTK mRNAs expressed in a freshly isolated TSC preparation. We identified 10 different PTK cDNAs among the 216 cDNAs sequenced, and demonstrate that transcripts of three of those (ufo, fyn and fer) are widely expressed among a large panel of immortalized thymic epithelial cell lines (TEC) and in primary cultures of TSC. Of the other seven, none were expressed in established TEC lines but, instead, displayed distinct expression patterns in cell types likely to have contaminated the fresh TSC preparation, i.e., macrophages, B cells, T cells and fibroblasts. Among the three PTK expressed in TEC lines, only one, ufo, exhibited expression exclusively in cells of non-hemopoietic origin. Although expression of ufo (also known as tyro 7, axl or ark) is not thymic-specific, in that it is also expressed in cell types of mesodermal origin in other tissues, its presence in TEC suggests a role for ufo in differentiation of the TSC compartment. Consistent with this notion, high-level expression of this receptor PTK at the protein level could be documented in every TEC line investigated, as well as in fresh thymus tissue sections. These data provide the first example of a receptor PTK in TSC and open new approaches to study the regulation of TSC differentiation.

Isolation of an immunodominant viral peptide that is endogenously bound to the stress protein GP96/GRP94.

Heat shock protein gp96 primes class I restricted cytotoxic T cells against antigens present in the cells from which it was isolated. Moreover, gp96 derived from certain tumors functions as an effective vaccine, causing complete tumor regressions in in vivo tumor challenge protocols. Because tumor-derived gp96 did not differ from gp96 isolated from normal tissues, a role for gp96 as a peptide carrier has been proposed. To test this hypothesis, we analyzed whether such an association of antigenic peptides with gp96 occurs in a well-defined viral model system. Here we present the full characterization of an antigenic peptide that endogenously associates with the stress protein gp96 in cells infected with vesicular stomatitis virus (VSV). This peptide is identical to the immunodominant peptide of VSV, which is also naturally presented by H-2Kb major histocompatibility complex class I molecules. This peptide associates with gp96 in VSV-infected cells regardless of the major histocompatibility com- plex haplotype of the cell. Our observations provide a biochemical basis for the vaccine function of gp96.