The minimal number of class II MHC-antigen complexes needed for T cell activation.

Major histocompatibility complex (MHC) molecules are exposed to large quantities of self and nonself antigens. It is not known what fraction of MHC molecules needs to be occupied by antigen to induce a T cell response. A quantitative study of naturally processed antigen indicated that T cells could be activated when only 0.03 percent of the total I-Ed purified from antigen-presenting cells (APCs) was occupied with antigen. B cells and macrophages processed hen egg lysozyme (HEL) with different efficiencies, but similar degrees of occupancy were required for T cell stimulation. Higher occupancy was needed for I-Ed-transfected L cells, possibly reflecting the requirement for other accessory molecules for efficient APC-T cell interaction.

Epitope mapping by chemical modification of free and antibody-bound protein antigen.

A monoclonal antibody bound to a protein antigen slows the rate of chemical modification of amino acid residues located at the epitope. By comparing the degree of acetylation of 18 lysine and 7 threonine residues in free and antibody-bound horse cytochrome c, a discontiguous, conformational epitope was characterized on this protein antigen. The new approach is particularly suitable to probe discontiguous and conformational epitopes, which are difficult to analyze by other procedures.

Processing of DR1-restricted determinants from the fusion protein of measles virus following two distinct pathways.

A panel of human T cell clones specific for measles virus was characterized and among them fusion protein-specific, DR1-and DP-restricted T cell clones were selected to study the processing and presentation of determinants borne by a viral membrane protein. Using two independent methods to assess the activation of T cells when they encounter antigen-presenting cells, proliferation assay and Ca2+ flux measure by flow cytometry, we show that determinants from the fusion protein of measles virus presented to two DR1-restricted T cell clones have strikingly different processing requirements. While treatment with chloroquine, leupeptin and brefeldin A of antigen-presenting cells infected with the measles virus inhibits presentation of the first determinant, presentation of the second is prevented only by leupeptin but not by chloroquine and brefeldin A. The major histocompatibility complex deletion mutant cell line T2 was transfected with DR alpha and DR1 beta genes to be tested as antigen-presenting cells with the measles virus-specific T cell clones. DR1-transfected T2 cells infected with the measles virus presented the fusion protein determinant whose processing was sensitive to chloroquine and brefeldin A but failed to display insensitivity to these two drugs, further indicating that the two determinants are generated following two distinct pathways. The first is likely to be independent of the expression of the class II major histocompatibility complex-like molecule DM, whereas the other requires it. In conclusion, determinants on the same polypeptide can have profoundly dissimilar processing requirements. Due to transport to successive compartments with different processing capabilities, more determinants are successfully released from antigens and/or captured by class II major histocompatibility complex molecules, thereby increasing the repertoire of determinants displayed by class II major histocompatibility complex molecules.

Release of DR molecules from complexes with invariant chain through the formation of a C-terminal 25 kDa invariant chain fragment.

To investigate how class II major histocompatibility complex (MHC) molecules are released from complexes with invariant chain (Ii), we studied a 25 kDa Ii fragment (p25) detected by Western blotting in affinity chromatographed DR preparations. The p25 species corresponds to the non-transmembrane, C-terminal Ii fragment 107-232. It was determined by gel filtration chromatography that the p25 fragment has a relative molecular mass (M(r)) of 46 kDa, indicating that this Ii fragment is present as dimers in B cell lysate. Two independent approaches were followed to demonstrate that generation of the p25 fragment takes place shortly before, or concomitantly to, loading of class II MHC molecules with antigen fragments. First, it was shown that a fraction of the p25 molecules is resistant to endoglycosidase H digestion, indicating that the p25 polypeptide can exit the endoplasmic reticulum (ER) and is transported at least to the cis-Golgi compartment. Second, treatment of class II MHC-positive B cells with leupeptin blocks the formation of p25, further indicating that this Ii fragment is generated in the endosomal compartment. The role of the p25 Ii species in the assembly of complexes between peptides and DR molecules was then investigated. While the p25 fragment was totally unable to prevent binding of a synthetic tetanus toxin peptide to DR molecules, the full-length Ii species (p33/35) effectively inhibited peptide binding, indicating that, by contrast with the p33/35 species, the p25 fragment does not occlude the peptide binding site of DR molecules. We concluded that the p25 fragment, which is produced by proteolytic cleavage at the N-terminal side of Methionine 107, has a decreased affinity for DR molecules as compared with the p33/35 species. Dissociation of the p25 fragment from DR molecules exposes the peptide binding site, which is thus made accessible for antigen fragments. This model of the complexes between DR and antigen fragments proposes that a stretch of Ii prevents peptide binding by occluding the peptide binding site without directly occupying it.

Inhibition of peptide binding to DR molecules by a leupeptin-induced invariant chain fragment.

Loading of peptides onto DR molecules was studied by characterizing precursors of the mature peptide-DR complexes expressed at the surface of B cells. Since invariant chain (Ii) prevents binding of peptides by DR molecules, it was speculated that analysis of complexes between DR heterodimers and proteolytic fragments of Ii offers the possibility to examine how DR molecules and peptides assemble. Using a procedure combining a two-step affinity chromatography and gel filtration, we isolated from leupeptin-treated B cells complexes between DR molecules and N-terminal Ii fragments previously called "leupeptin-induced polypeptides" (LIP; Blum and Cresswell, 1988, Proc. natn. Acad. Sci. U.S.A. 85, 3975-3979). It was observed that the most prominent LIP fragment has a relative molecular mass (M(r)) of 16 kDa. In addition, we show that this polypeptide species does not bear N-linked glycans, indicating that this fragment does not extend beyond residue 129 of Ii. Similarly to DR alpha beta heterodimers associated with the full length 33 and 35 kDa Ii forms, DR alpha beta heterodimers associated with LIP fragments are unstable in sodium dodecyl sulfate (SDS) at ambient temperature, whereas mature DR alpha beta heterodimers are resistant to dissociation with SDS. These results are indirect evidence that LIP-DR complexes are devoid of bound peptides. This possibility was supported by showing that LIP-DR complexes fail to bind a radioiodinated tetanus toxin peptide (125I-p2), while DR molecules, which are spontaneously released from complexes with LIP fragments, bind the labeled peptide. These results demonstrate that association with LIP fragments is sufficient to prevent binding of peptides by DR molecules. This notion was further documented by showing that binding of 125I-p2 on DR heterodimers is inhibited by preparations of LIP fragment. By contrast, a soluble recombinant fragment corresponding to the extracytoplasmic region of Ii did not block 125I-p2 binding. The results presented in this study indicate that the cytoplasmic and/or transmembrane region of Ii is required to prevent peptide binding by DR molecules, while the extracytoplasmic portion of Ii, though capable of associating with DR molecules, lacks the capacity to block peptide binding.

Native-like, long synthetic peptides as components of sub-unit vaccines: practical and theoretical considerations for their use in humans.

Vaccines have been used as a successful tool in medicine by way of controlling many major diseases. In spite of this, vaccines today represent only a handful of all infectious diseases. Therefore, there is a pressing demand for improvements of existing vaccines with particular reference to higher efficacy and undisputed safety profiles. To this effect, as an alternative to available vaccine technologies, there has been a drive to develop vaccine candidate polypeptides by chemical synthesis. In our laboratory, we have recently developed a technology to manufacture long synthetic peptides of up to 130 residues, which are correctly folded and biologically active. This paper discusses the advantages of the molecularly defined, long synthetic peptide approach in the context of vaccine design, development and use in human vaccination.

A novel and simple procedure for determining T cell epitopes in protein antigens.

A human T cell epitope was mapped on tetanus toxin by the use of a simple method. Tetanus toxin peptide preparations were produced by various specific chemical and proteolytic cleavages. T cell proliferative response to these peptide preparations was assayed to determine which amino acid residues were present or absent in the T cell epitope. Cyanogen bromide-derived tetanus toxin peptides were separated by electrophoresis and transferred onto nitrocellulose. T cell proliferative response to nitrocellulose fragments was assayed to determine the molecular weight of the antigenic tetanus toxin peptide. The combination of the two procedures allowed us to define a 21-mer tetanus toxin peptide recognized by the T cell clone. This method represents an easily performed alternative for the mapping of T cell epitopes on protein antigen.

Differential antigenicity of recombinant polyepitope-antigens based on loop- and helix-forming B and T cell epitopes.

To investigate a strategy for the design of chimeric antigens based on B cell epitopes (BCEs) we have genetically recombined multiple copies of loop- (L) and helix-forming (H) sequential and protective BCEs of the measles virus hemagglutinin protein (MVH) in a number of high-molecular-weight polyepitope constructs (24.5-45.5 kDa). The BCE cassettes were combined semi-randomly together with a promiscuous T cell epitope (TCE; tt830-844) to yield 13 different permutational constructs. When expressed in mammalian cells, all constructs were detectable by Western blot as distinct bands of predicted molecular weight. Flow cytometry with conformation-specific antibodies revealed the Cys-loop in two [(L(4)T(4))(2) and (L(2)T(2))(4)] and the helix conformation in one [(H(2)T(2))(4)] of the different permutational constructs. The larger constructs, containing 16 epitope cassettes, seemed more likely to express the BCEs in their native conformation than the 8-mers. In the T cell proliferation assay, constructs with a higher copy number of TCEs, such as (L(2)T(2))(4), were more antigenic, as long as tandem repeats were separated by spacers. Since the conformation of even sequential BCEs and the processing of TCEs are both sensitive to their molecular environment it is difficult to predict the antigenic properties of polyepitopes. However, with the permutational approach we have developed several polyepitope constructs [(L(4)T(4))(2), (L(2)T(2))(4), (H(2)T(2))(4)] based on complex sequential BCEs that are antigenic for both T and B cells. Several constructs induced sera that reacted with reporter peptides, demonstrating that the sequential nature of the viral epitopes was conserved in the polyepitopes. Although several sera contained antibodies directed against amino acids critical for neutralization, only one construct induced antibodies that cross-reacted with the virus. Our results show the difficulty of designing chimeric antigens based on B cell epitopes mimicking their antigenic and immunologic properties even when these are sequential in nature.

Peptide-MHC complexes assembled following multiple pathways: an opportunity for the design of vaccines and therapeutic molecules.

Antigen degradation and peptide loading to major histocompatibility complex class I and class II molecules are described with special emphasis on "noncanonical" pathways. Examples of specific peptide loading for measles proteins are provided. In addition, characterization of defined epitopes presented to T cells can lead to the design of products of special interest in medicine and, in particular, in development of vaccines.

Antigen presenting cells: detection and quantification of a cytochrome c determinant important for activation of T-cells on bone marrow derived macrophages by using specific anti cytochrome c monoclonal antibody.

The region of the horse cytochrome c molecule recognized by Mab SJL2-4 specific for the denatured form of the protein was located around residues 22-28. Binding studies on antigen pulsed macrophages were also performed. Surprisingly, heme peptide 1-65 was not recognised by Mab when bound on macrophages. This correlates with the incapacity of the same peptide to activate the T-cell clone 2-16. Binding sites on antigen pulsed macrophages varied between 0.5-2 x 10(6) per cell depending on the conditions used. The expression of the antigenic determinant as detected by Mab was also followed under different conditions (chloroquine, trypsin treatment) and time. Kinetics parameters of the antigen-antibody reaction in solution and on antigen bound macrophages were also determined and are dramatically different. This is correlated with a different structure of the peptide in solution and on macrophage cell surface.

A novel potent Fas agonist for selective depletion of tumor cells in hematopoietic transplants.

There remains a clear need for effective tumor cell purging in autologous stem cell transplantation (ASCT) where residual malignant cells within the autograft contribute to disease relapse. Here we propose the use of a novel Fas agonist with potent pro-apoptotic activity, termed MegaFasL, as an effective ex-vivo purging agent. MegaFasL selectively kills hematological cancer cells from lymphomas and leukemias and prevents tumor development at concentrations that do not reduce the functional capacity of human hematopoietic stem/progenitor cells both in in vitro and in in vivo transplantation models. These findings highlight the potential use of MegaFasL as an ex-vivo purging agent in ASCT.

DR alpha beta dimers released from complexes with invariant chain fail to stimulate alloreactive T cell clones.

To demonstrate that DR alpha beta dimers still complexed to invariant chain (Ii) have not yet acquired peptides recognized by alloreactive T cells, complexes between DR molecules and Ii isolated from Epstein-Barr-virus (EBV)-transformed B cells were analyzed by affinity chromatography and gel filtration. First, it was shown that DR/Ii complexes inserted into artificial planar membranes (PM) failed to stimulate proliferative response of five alloreactive T cell clones and a polyclonal alloreactive T cell line, while PM bearing mature DR alpha beta dimers from the same EBV-B cells were stimulatory for the T cell clones and the T cell line. These findings indicate that either Ii inhibits binding of peptides to DR molecules or Ii hinders T cells recognition of peptide/DR complexes. Second, to discriminate between these two possibilities, DR alpha beta dimers, which were artificially released from complexes between DR molecules and Ii, were inserted into PM. These DR alpha beta dimers were devoid of alloreactive stimulatory capacity while fully capable of binding and presenting a tetanus toxin synthetic peptide to a specific T cell clone, indicating that DR molecules released from complexes with Ii are empty. This study, by showing that DR molecules bound to Ii do not bear peptides recognized by alloreactive T cells, supports the notion that association of Ii with class II major histocompatibility complex (MHC) molecules prevents premature peptide loading and hence favors encounter with peptides derived from proteins of the extracellular compartment. Since allogeneic class II MHC molecules released from complexes with Ii were not stimulatory for five out of five alloreactive T cell clones and a polyclonal alloreactive T cell line, these data also indicate that, in most cases, alloreactive T cells recognize ligands constituted by complexes between allogeneic class II MHC molecules and specific peptides which derive from the antigen-presenting cells themselves or serum proteins.

The relative abundance of two measles virus fusion protein peptide-DR1 complexes expressed by B cells is independent of the form of the antigen.

The relative processing and presentation efficiency of two DR1-restricted determinants from the fusion protein (F protein) of measles virus (MV) was determined using three forms of antigen (Ag): MV, an F protein recombinant vaccinia virus, and a chimerical polypeptide between the glutathione S-transferase and the F protein (GST-F protein). First, it was shown that these different preparations of F protein have distinct processing requirements. In MV-infected B cells, the F254 determinant (contained within the F protein sequence 254-268) relies on protein synthesis for its presentation, while the F314 determinant (contained within the F protein sequence 314-328) is also presented in the absence of protein synthesis. By contrast, in GST-F protein-pulsed B cells, presentation of both determinants is dependent on protein synthesis. Then, it was established that, independently of the form of the Ag, the F314 determinant was considerably more (18- to 36-fold) efficiently processed and presented than the F254 determinant. These results indicate that determinants from the same protein are displayed by antigen-presenting cells at widely different levels and they may also suggest that this is an intrinsic characteristic of the determinants, rather than a feature controlled by the processing pathways followed by the Ag.

Characterization of a naturally processed MHC class II-restricted T-cell determinant of hen egg lysozyme.

Compelling evidence indicates that T cells recognize complexes formed by major histocompatibility complex-encoded molecules and antigenic peptide fragments. This is based largely on the ability of small synthetic peptides to substitute for naturally processed antigen in stimulating T cells. Naturally processed fragments of exogenous antigen are thought to arise by limited proteolytic degradation of native antigen inside acidic compartments of antigen-presenting cells, but until now no physiologically processed antigen has been directly analysed. Here we report the characterization of physiologically processed antigen eluted from mouse class II major histocompatibility complex I-Ed molecules. The antigenic material corresponds to a previously described antigenic determinant of hen egg lysozyme (HEL 107-116) and has a relative molecular mass Mr of about 2,000. HPLC analysis identified at least two or three separate molecular species, suggesting limited, albeit significant, heterogeneity of naturally processed peptides. Finally, under our experimental conditions, it was calculated that a substantial proportion (10-40%) of I-Ed molecules were occupied by these HEL-derived antigenic determinants.

Kinetics of MHC-antigen complex formation on antigen-presenting cells.

With the use of flow cytometry, we recorded changes in intracellular ionized calcium [Ca2+]i of Indo-1 loaded T cells that were triggered by contact with APC. This rapid readout of TCR perturbation enabled us to monitor the formation of stimulatory Ag-MHC complexes on EBV-transformed B cells that were either pulsed with native tetanus toxoid (TT) or with a 12-amino-acid fragment of this protein. Neither unpulsed APC nor Ag-specific APC that were pulsed with native Ag and kept at +4 degrees C were able to induce changes in basal T cell [Ca2+]i in TT-specific T cell clones. After 1 h at 37 degrees C, however, the Ag-pulsed APC were able to induce a three-to-fourfold increase in [Ca2+]i. This length of time appeared to be almost independent of the concentration of Ag with which the APC were pulsed, suggesting that the lag time was due more to intracellular transit than to association of the processed Ag with the MHC molecule. Furthermore, the same lag time and independence of Ag concentration were found when the EBV-transformed B cells were pulsed with a mouse-anti-transferrin receptor mAb and tested for their capacity to trigger a T cell clone specific for processed mouse Ig. This indicates that, in addition to surface Ig, other receptors that are internalized can function in the same fashion in the uptake and processing of a soluble Ag. In contrast to what was found with intact native Ag, no lag time was observed when the APC were pulsed with high concentrations of a 12-amino-acid peptide, containing the amino acid sequence recognized by a TT-specific T cell clone, suggesting that the formation of MHC-peptide complexes occurs instantly. Pulsing with a lower peptide concentration, however, caused the appearance of a time-dependent increase in efficacy of Ag presentation, suggesting a slow accumulation of MHC-peptide complexes on the B cell membrane.

Cell type-specific processing of the I-Ed-restricted hen egg lysozyme determinant 107-116.

Class II major histocompatibility complex heterodimers present to T cells determinants as sets of antigen fragments with ragged N and C termini. It is not yet elucidated whether different types of antigen-presenting cells generate identical sets of peptides containing the same determinant. Taking advantage of recombinant I-Ed molecules produced by insect cells as empty heterodimers, a sensitive T cell stimulation assay was developed to analyze naturally processed hen egg lysozyme (HEL) peptides. I-Ed preparations were isolated from antigen-presenting cells cultured with HEL. Following acid treatment, peptides eluted from I-Ed were chromatographed and the fractions incubated at acidic pH with purified recombinant I-Ed molecules, conditions which favor peptide binding. The stimulatory capacity of the reconstituted peptide-I-Ed complexes adsorbed on the well surface of cell culture plates was then evaluated by measuring interleukin-2 secreted by an HEL 107-116-specific, I-Ed-restricted T cell hybridoma. We found that the B lymphoma A20 and an I-Ed-transfected fibroblast cell line generated distinct sets of peptides containing the HEL sequence 107-116. Our results suggest the possibility that presentation of one determinant by different types of antigen-presenting cells stimulates populations of T cells with distinct fine antigen specificities.

Kinetics and extent of protein tyrosine kinase activation in individual T cells upon antigenic stimulation.

Using human CD4+ T-cell clones and peptide-pulsed antigen-presenting cells (APC) we measured, at the single cell level, different steps in the T-cell activation cascade. Simultaneous analysis of T-cell antigen receptor (TCR) down-regulation and interferon-gamma (IFN-gamma) production shows that both the level of TCR occupancy and the amount of IFN-gamma produced by single T cells increase in an antigen dose-dependent fashion. Conversely, commitment of T cells to IFN-gamma production does not occur as soon as a defined number of TCR have been engaged, but requires the same duration of sustained signalling at low as well as at high antigen concentrations. Measurement of phosphotyrosine levels by flow cytometry reveals that, upon conjugation with APC, individual T cells undergo an antigen dose-dependent activation of protein tyrosine kinases (PTK), which parallels the level of TCR occupancy. In antigen-stimulated T cells the increased phosphotyrosine staining is localized in the area of contact with APC, as shown by confocal microscopy. PTK activation is sustained for at least 2 hr after conjugation, and is required to maintain a sustained increase in intracellular Ca2+ concentration. Our results show, for the first time, a direct correlation between the level of TCR occupancy and the activation of PTK in individual T cells and offer an explanation for how the number of triggered TCR can be 'counted' and integrated in a corresponding biological response.

Universally immunogenic T cell epitopes: promiscuous binding to human MHC class II and promiscuous recognition by T cells.

To understand the effect of human MHC class II polymorphism on antigen recognition, we analyzed the memory T cell response to three tetanus toxin epitopes defined by three short synthetic peptides (p2, p4 and p30). We found that p2 and p30 are universally immunogenic, since they are recognized by all primed donors, irrespective of their MHC haplotypes. The analysis of specific clones indicates that both peptides are very promiscuous in their capacity to bind to class II. p30 can be recognized in association with DRw11(5), 7, 9 and with DPw2 and DPw4, while p2 can be recognized in association with DR1, DRw15(2), DRw18 (3), DR4Dw4, DRw11(5), DRw13(w6), DR7, DRw8, DR9, DRw52a and DRw52b. On the contrary, the third peptide, p4, can be recognized by only half of the donors in association with only DRw52a and DRw52c. Analysis of truncated peptides shows that p30 contains three distinct epitopes, each recognized in association with different class II molecules. Therefore, the restriction specificity is already set at the level of the peptide-MHC complex and, in all cases, T cells discriminate p30 bound to different class II molecules. On the contrary, p2 contains only one epitope, which is recognized in association with all DR molecules. In this case we found two different restriction patterns. Some clones are monogamous, since they recognize the peptide in association with one DR allele, while others are promiscuous, since they recognize by peptide in association with several different DR molecules. Thus, in this case, the restriction specificity is also set at the level of the T cell receptor. We suggest that both the promiscuous binding of peptides and the promiscuous recognition by T cells are dependent on the particular structure of the DR molecules, having a monomorphic alpha chain associated with a polymorphic beta chain.

Processing of tetanus toxin by human antigen-presenting cells. Evidence for donor and epitope-specific processing pathways.

Human T cell clones specific for epitopes 830-843 and 947-967 of tetanus toxin can be differentially activated in vitro when APC (PBL or LCL) from different donors are pulsed with tetanus toxin. Although PBL tested do not seem to exhibit substantial differences in the number of precursor T cells specific for these epitopes, APC from the same donors activate clone KT-2 specific for peptide 830-843 but not clone KT-30 specific for peptide 947-967. These APC express the proper restriction element because they can present the corresponding synthetic peptides. The failure to present a particular epitope might, however, be explained by the absence or presence of a protease(s) required for Ag presentation that may vary for different epitopes. Indeed, the protease inhibitor leupeptin was found to inhibit activation of KT-2 but not KT-30 T cell clone by the KK.35 B cell line normally capable of presenting either epitope. In summary, these data suggest that tetanus toxin processing and epitope formation by APC is distinct in different donors and for different epitopes.