Peptide length and sequence specificity of the mouse TAP1/TAP2 translocator.

The transporter associated with antigen processing (TAP) delivers peptides to the lumen of the endoplasmic reticulum in an adenosine triphosphate (ATP) dependent fashion for presentation by major histocompatibility complex class I molecules. We show that the mouse TAP translocator (H-2b haplotype) selects peptides based on a minimal size of nine residues, and on the presence of a hydrophobic COOH-terminal amino acid. The preponderance of COOH-terminal hydrophobic amino acids in peptides capable of binding to mouse class I molecules thus fits remarkably well with the specificity of the TAP translocator. In addition to transport in the lumenal direction, efflux of peptide in the cytosolic direction is observed in an ATP- and temperature-dependent manner. By maintaining a low peptide concentration at the site of class I assembly, this efflux mechanism may ensure that class I molecules are loaded preferentially with high affinity peptides.

Linear differentiation of cytotoxic effectors into memory T lymphocytes.

A central question in immunology is the origin of long-lived T cell memory that confers protection against recurrent infection. The differentiation of naïve T cell receptor transgenic CD8+ cells into effector cytotoxic T lymphocytes (CTLs) and memory CD8+ cells was studied. Memory CD8+ cells that were generated after strong antigenic stimulation were the progeny of cytotoxic effectors and retained antigen-specific cytolytic activity 10 weeks after adoptive transfer to antigen-free recipient mice. Thus, potential vaccines based on CTL memory will require the differentiation of naïve cells into post-effector memory T cells.

MCC, a candidate familial polyposis gene in 5q.21, shows frequent allele loss in colorectal and lung cancer.

MCC is a gene located within human chromosome band 5q.21 that shows somatically acquired mutations in colorectal cancer, and may be identical to the gene responsible for inheritance of familial adenomatous polyposis. Here we demonstrate that alleles contiguous with or within MCC are deleted in a high proportion of sporadic colorectal carcinomas. Of 106 carcinomas that were informative concurrently at close-flanking sites both centromeric and telomeric to MCC, 41.5% showed acquired allele loss contiguous with MCC. Evidence is presented to show that the true frequency of loss of MCC alleles is higher still. In contrast, allele losses in chromosome 5 that were incompatible with involvement of MCC were very rare (2% of a total series of 201 informative tumours). Interstitial deletion was the commonest mechanism of allele loss, and L5.71-3, a probe known to include coding sequences of MCC, marks the most consistently deleted site. Moreover mapping of chromosome breakpoints with six probes within 5q.21 sited the common critical deletion in a 2.5 Mb region which included L5.71-3. However use of L5.71-3 itself suggested that critical deleted regions may lie on either side of the probed sequence. The simplest explanation for this unexpected finding is that MCC itself is the essential deleted gene, the lost exons lying sometimes centromeric to, sometimes telomeric to and occasionally within the region detected by L5.71-3. Tumours in which MCC-related alleles were lost by interstitial deletion were in general larger than those with other mechanisms of acquired homozygosity (e.g. mitotic recombination), but there were no other obvious associations with clinicopathological features. Between 20% and 25% of lung cancers also showed acquired allele losses contiguous with MCC. The significance of this observation is still to be determined, as lung tumours show allele losses at many other sites, but the specificity of the probes used in this study does establish that the 5q.21 losses in these tumours are compatible with involvement of MCC.

High frequency of APC loss in sporadic colorectal carcinoma due to breaks clustered in 5q21-22.

Familial adenomatous polyposis is transmitted by a gene (APC) located within 5q21-22. Hemizygous loss of at least a part of 5q has been reported in 19-36% of sporadic colorectal carcinomas. This suggests that an anti-oncogene is located on that chromosome arm, but the probes used previously gave little information on the status of APC in the tumours. Using DNA probes homologous to polymorphic sequences flanking and close to the APC locus we show that more than half of a large series of carcinomas had lost at least one flanking allele. Mapping of allele losses provides data that imply clustering of breakpoints in a 10-15 megabase region around APC. The commonest chromosome defect responsible for APC loss was interstitial deletion. Mitotic recombination or partial arm loss were less frequent mechanisms. Whole chromosome loss was rare. This pattern contrasts with that reported in acquired homozygosity at other anti-oncogene loci in sporadic tumours and implies that APC loss is an early event in colorectal carcinogenesis. This view is also supported by the observations that 5q21-22 loss occurs with similar frequency in DNA diploid and DNA aneuploid tumours, and also in tumours at all clinical stages of progression.

The role of peptide in the positive selection of CD8+ T cells in the thymus.

Recent experiments using fetal thymus organ culture (FTOC) systems derived from class I deficient mice have provided evidence supporting the view that specific recognition of class I MHC plus peptide complexes is required for the positive selection of CD8+ T cells. The addition of class I binding peptides to FTOC systems derived from beta 2-microglobulin and TAP 1 deficient mice induced the positive selection of CD8+ T cells in a peptide specific manner. The specific recognition of peptide during positive selection implies that the repertoire of specificity's exhibited by CD8+ T cells is at least in part determined by the repertoire of self-peptides presented by class I MHC in the thymus.

A differential-avidity model for T-cell selection.

The processes of positive and negative selection during thymic development shape the repertoires of antigen specificities displayed by T cells. This rids the animal of potentially autoreactive T cells and, at the same time, ensures that they are capable of major histocompatibility complex (MHC)-restricted recognition of antigen. Paradoxically, both processes involve the engagement of the T-cell recepetor (TCR) on immature thymocytes with peptide/MHC complexes expressed on thymic stromal cells. Here, Philip Ashton-Rickardt and Susumu Tonegawa suggest that the critical parameter determining the outcome of this interaction is the number of TCRs occupied by peptide/MHC complexes and that this, in turn, is determined by the avidity of the TCR-MHC interaction: low avidity resulting in positive selection and high avidity resulting in negative selection.

Mutants of the hepatitis B virus surface antigen that define some antigenically essential residues in the immunodominant a region.

The nature of the immunodominant a region of hepatitis B virus surface antigen (HBsAg) has been examined by mutation of specific amino acid residues. Proline 142 was required for the exhibition of full antigenicity. Replacement of cysteines 124 and 147 by serines drastically reduced or eliminated reactivity with antibodies to HBsAg, which implicates these two residues in stabilising the conformation of the antigen. The a region of HBsAg has also been shown to influence both the immunoreactivity of the adjacent subtype antigenic region, despite being immunologically distinct from it, and the ability of the antigen to interact with a subtype-specific monoclonal antibody. These results emphasise the importance of the polypeptide region between cysteine residues 124 and 147 in determining a antigenicity, as well as manifestation of the subtype of HBsAg, and reinforce the view that these are conformational epitopes.

Mutations that change the immunological subtype of hepatitis B virus surface antigen and distinguish between antigenic and immunogenic determination.

The molecular basis of the d or y immunological subtype of hepatitis B virus (HBV) surface antigen (HBsAg) has been investigated by mutation of specific amino acid residues. When combined with substitution of serine 113 by threonine, replacement of arginine 122 by lysine or of tyrosine 134 by phenylalanine, or both of these changes, altered the antigenic subtype of HBsAg from y+d- to y+d+. These same mutations had a more dramatic effect on the subtype of antibodies induced by the antigens, a combination of all three mutations completely changing the subtype from y to d. Our study thus identifies residues in HBsAg that not only affect the subtype but discriminate between changes in antigenic and immunogenic behaviour. It also shows how the y and d subtypes may be manifest by the same molecule.

Genetic engineering applied to the development of vaccines.

The simplest application of the modern genetic manipulation methods to vaccine development is the expression in microbial cells of genes from pathogens that encode surface antigens capable of inducing neutralizing antibodies in the host of the pathogen involved. This procedure has been exploited successfully for development of a vaccine against hepatitis B virus (HBV) that is now widely used. Similar approaches have been directed towards formulations for immunization against several other animal and human diseases and some of these preparations are now presently in trials. Of no less importance is the impact of biotechnology in providing reagents for fundamental studies of topics such as the determination of virulence, antigenic variation, virus receptors and the immunological response to viral antigens. The core antigen of HBV is a good example of a product of genetic engineering that is a valuable diagnostic reagent, and that is finding important use in immunological studies of particular pertinence to vaccine development.

Suicide induced by cytolytic activity controls the differentiation of memory CD8(+) T lymphocytes.

Cytotoxic T lymphocytes (CTL) confer protection against intracellular pathogens, yet the mechanism by which some escape activation induced cell death (AICD) and give rise to long-lived memory cells is unclear. We studied the differentiation of transgenic TCR CD8(+) cells into CTL and memory cells using a novel system that allowed us to control cytolytic activity. The perforin/granzyme granules used to lyse targets induced the apoptosis of CTL in a fratricide-independent manner. After adoptive transfer to antigen-free mice, the ability of CTL to give generate memory cells was determined. We found that the extent of cytolysis by a common pool of CTL controlled the differentiation into memory cells, which were only generated under conditions of minimal cytolytic activity. Thus, the differentiation of naive CD8(+) cells into memory cells may not depend on the presence on a subset of committed CTL precursors, but rather is controlled by the extent of granule-mediated cytolysis.

Limits to the differential avidity model of T cell selection in the thymus.

It has been postulated that the critical feature that determines the developmental fate of an immature thymocyte is the avidity of interaction between thymocyte TCR and peptide/MHC molecules on thymic stromal cells. However, it is possible that certain innate properties of peptides predispose them to triggering only positive or negative selection irrespective of their density on thymic stromal cells. To distinguish between these hypotheses, we examined the ability of several different peptides to induce the positive and negative selection of TCR transgenic (P14) antilymphocytic choriomeningitis virus (LCMV) CTLs in fetal thymus organ cultures (FTOC) from TAP1+ and TAP1- mice. We found that only relatively weak agonist peptides could induce the positive selection of anti-LCMV CTLs. A nonagonist peptide could induce positive selection but not negative selection; however, a weak agonist peptide could induce the positive selection of anti-LCMV CTLs in P14 TAP1- FTOC and negative selection in P14 TAP1+ FTOC. These data imply that there are upper and lower limits for the affinity of a peptide in triggering positive or negative selection, but that for peptides of intermediate affinity the overall avidity of interaction with the P14 TCR is the critical parameter in determining the developmental fate of thymocytes. Our observations also suggest a prominent role for low affinity self peptides in selecting a function repertoire of CD8+ T cells.

TAP1 mutant mice are deficient in antigen presentation, surface class I molecules, and CD4-8+ T cells.

The transporter associated with the antigen processing 1 (TAP1) gene encodes a subunit for a transporter, presumed to be involved in the delivery of peptides across the endoplasmic reticulum membrane to class I molecules. We have generated mice with a disrupted TAP1 gene using embryonic stem cell technology. TAP1-deficient mice are defective in the stable assembly and intracellular transport of class I molecules and consequently show severely reduced levels of surface class I molecules. These properties are strikingly similar to those described for the TAP2 mutant cell line RMA-S. Cells from the TAP1-deficient mice are unable to present cytosolic antigens to class I-restricted cytotoxic T cells. As predicted from the near absence of class I surface expression, TAP1-deficient mice lack CD4-8+ T cells.

Presentation of endogenous viral proteins in association with major histocompatibility complex class II: on the role of intracellular compartmentalization, invariant chain and the TAP transporter system.

Major histocompatibility complex (MHC) class II-associated antigen presentation is mainly linked to processing of exogenous antigens upon cellular uptake by endocytosis, but has also been observed for endogenously synthesized antigens. We have studied the MHC class II-associated presentation of the endogenously synthesized membrane associated glycoprotein (GP) and the cytosolic nucleoprotein (NP) of lymphocytic choriomeningitis virus (LCMV) in professional antigen presenting cells (APC) of mice. Since LCMV is a noncytopathic virus and minimally affects cellular protein synthesis, it is a convenient virus for the study of antigen presentation. In contrast, most other studies assessing class II-associated presentation of endogeneously synthesized viral antigens used cytolytic viruses such as vaccinia, measles and influenza virus, which drastically interfere with host cell functions. In addition, most studies were performed using non-professional APC. We found that class II-associated presentation of endogenously synthesized membrane associated LCMV-GP was efficient and could not be inhibited by chloroquine or leupeptin. Neither the transporter associated with processing (TAP) system nor the invariant chain (Ii) were significantly involved in this process. In contrast, MHC class II-associated presentation of endogenously synthesized cytosolic LCMV-NP was not observed even in Ii-deficient APC. Thus, MHC class II loading of endogenously synthesized LCMV-GP apparently does not require processing in acidic endosomal compartments as defined by chloroquine and leupeptin insensitivity. Furthermore, although the TAP molecules transport peptides of up to 15 amino acids in length, which potentially could bind to MHC class II molecules in the endoplasmic reticulum, such a process apparently does not occur for either the glycoprotein or the nucleoprotein. Therefore, the subcellular localization of an endogenously synthesized protein influences crucially whether or not MHC class II loading can occur independently of the acidic compartments usually involved in MHC class II loading.

Positive selection of self- and alloreactive CD8+ T cells in Tap-1 mutant mice.

Mice with a homozygous deletion in their Tap-1 gene (-/- mice) express very low levels of cell membrane major histocompatibility complex class I molecules and have < 1% peripheral CD8+ T cells. We show that these -/- mice but not their +/- littermates display strong primary syngeneic anti-H-2Kb and -Db-specific responses mediated by CD8+ T cells. These responses are augmented by in vivo priming. Further, -/- mice primed in vivo with H-2d alloantigens generate an anti-H-2d response which appears nearly as strong as that found in +/- littermates. Both -/- anti-H-2b and anti-H-2d T cells do not recognize target cells from Tap-1 -/- animals or Tap-2-deficient RMA-S cells. Thus, some CD8+ anti-self and alloreactive T cells can be selected in the absence of Tap proteins.

Peptide contributes to the specificity of positive selection of CD8+ T cells in the thymus.

Mice deficient in the gene encoding the peptide transporter associated with antigen processing (TAP1) have drastically reduced levels of surface expression of MHC class I molecules and few CD8+ T cells. Addition of class I binding peptides to cultured fetal thymi from TAP1 mutant mice invariably allowed the rescue of class I expression, while only some of these peptides promoted the positive selection of CD8+ T cells, which were polyclonal and specific for the peptide-MHC complex. A nonselecting peptide was converted to a mixture of selecting peptides when the residues involved in the interaction with TCRs were altered. A mixture of self-peptides derived from C57BL/6 thymi induced positive selection of CD8+ T cells at concentrations that gave relatively low class I surface expression. The implication of these observations is that self-peptides determine, in part, the repertoire of specificities exhibited by CD8+ T cells.

Differential reactivity of residual CD8+ T lymphocytes in TAP1 and beta 2-microglobulin mutant mice.

TAP1 -/- and beta 2-microglobulin (beta 2m) -/- mice (H-2b background) express very low levels of major histocompatibility complex (MHC) class I molecules on the cell surface. Consequently these mice have low numbers of mature CD8+ T lymphocytes. However, TAP1 -/- mice have significantly higher numbers of CD8+ T cells than beta 2m -/- mice. Alloreactive CD8+ cytotoxic T lymphocyte (CTL) responses were also stronger in TAP1 -/- mice than in beta 2m -/- mice. Alloreactive CTL generated in TAP1 -/- and beta 2m -/- mice cross-react with H-2b-expressing cells. Surprisingly, such cross-reactivity was stronger with alloreactive CTL from beta 2m -/- mice than with similar cells from TAP1 -/- mice. The beta 2m -/- mice also responded more strongly when primed with and tested against cells expressing normal levels of H-2b MHC class I molecules. Such H-2b-reactive CD8+ CTL from beta 2m -/- mice but not from TAP1 -/- mice also reacted with TAP1 -/- and TAP2-deficient RMA-S cells. In contrast, H-2b-reactive CD8+ CTL from neither beta 2m -/- mice nor TAP1 -/- mice killed beta 2m -/- cells. In line with these results, beta 2m -/- mice also responded when primed and tested against TAP1 -/- cells. We conclude that the reactivity of residual CD8+ T cells differs between TAP1 -/- and beta 2m -/- mice. The MHC class I-deficient phenotype of TAP1 -/- and beta 2m -/- mice is not equivalent: class I expression differs between the two mouse lines with regard to quality as well as quantity. We propose that the differences observed in numbers of CD8+ T cells, their ability to react with alloantigens and their cross-reactivity with normal H-2b class I are caused by differences in the expression of MHC class I ligands on selecting cells in the thymus.

TAP1-dependent peptide translocation in vitro is ATP dependent and peptide selective.

T cells detect infection of cells by recognizing peptide fragments of foreign proteins bound to class I molecules of the major histocompatibility complex (MHC) on the surface of the infected cell. MHC class I molecules bind peptide in the endoplasmic reticulum, and analysis of mutant cells has demonstrated that an adequate supply of peptides requires the presence of two genes in the MHC class II locus that encode proteins called transporters associated with antigen processing (TAP) 1 and 2. TAP1 and TAP2 are members of the ATP-binding cassette family of membrane translocators. In this study, we demonstrate in a cell-free system that TAP1 is part of an ATP-dependent, sequence-specific, peptide translocator.

Evidence for a differential avidity model of T cell selection in the thymus.

Positive and negative selection of a lymphocytic choriomeningitis virus (LCMV) peptide-specific, H-2Db-restricted T cell clone (P14) was studied using TAP1- and TAP1+ mice transgenic for P14 T cell receptor (TCR) alpha and beta genes. Positive selection of transgenic CD8+ P14 cells was impaired in TAP1- mice. Addition of the LCMV peptide to TAP1- fetal thymic organ cultures (FTOCs) at low and high concentrations induced positive and negative selection of CD8+ P14 cells, respectively, while addition of the same peptide to TAP1+ FTOCs induced negative selection even at low concentrations. Both types of selection were peptide specific. Thus, a critical parameter that controls the fate of a thymocyte seems to be the number of TCRs engaged with complexes of peptide and major histocompatibility complex. When this number is low, positive selection occurs, and when it is high, negative selection takes place. These findings support a differential avidity model of T cell selection.

Specific recognition of thymic self-peptides induces the positive selection of cytotoxic T lymphocytes.

To understand how thymic selection gives rise to T cells that are capable of major histocompatibility complex (MHC)-restricted recognition of antigen but are tolerant of self, we directly examined how peptide/MHC ligands expressed on thymic epithelial cells trigger the positive selection of immature thymocytes. We demonstrate that abundant self-peptides, purified from the H-2D(b) molecules of thymic epithelial cells, are specifically recognized during the positive selection of CD8+ T cells, implying that positive selection generates a repertoire of T cells that is weakly self-reactive. We also found that this recognition is somewhat cross-reactive, thereby providing an explanation for how the specific recognition of a limited repertoire of thymic self-peptides can select a diverse repertoire of T cells.