News and Publications

Recognition of Antigen and Superantigen by T-Cell Receptors

N.R.J. Gascoigne, S.M. Alam, C. Haarstad, C. Lin, M.V. McGuire, S. Redpath, B.-C. Sim, M.A. Zal, T. Zal, D. Lo,* P.J. Travers**

* Department of Immunology, TSRI
** University of London, London, England

Our group is interested in the structural basis of recognition by T cells during T-cell development and antigen recognition. The T-cell receptor (TCR) recognizes foreign antigens as fragments of protein presented on self-MHC proteins.

KINETICS OF THE TCR INTERACTION WITH AGONIST OR ANTAGONIST MHC-PEPTIDE LIGANDS

During development in the thymus, T cells are subjected to both positive and negative selection by similar ligands. Ligands that strongly activate mature T cells cause negative selection of immature thymocytes, whereas weaker ligands that act as partial agonists or antagonists in mature T cells cause positive selection of thymocytes. Using surface plasmon resonance, we previously determined the kinetics of binding of a series of MHC-peptide complexes to TCRs in solution. We found a 3-fold difference in affinity between the negative- and positive-selecting peptides, and the range from full activation to no activation covered about a 50-fold range in affinity. We have extended these studies to a natural ligand that causes positive selection but is too weak to act as an antagonist. This ligand has a measurable affinity that is lower than the affinities of the known antagonists but higher than those of completely null ligands.

Most interestingly, we have shown that strong agonists cause a change in the kinetics of TCR binding at 37°C but not at 25°C. Antagonist ligands do not show this effect. The half-lives of TCR complexes with agonist ligands are much greater than those of complexes with antagonist ligands. Analysis of the data suggests TCR dimerization: a single TCR molecule binds to the MHC-peptide complex and then recruits a second TCR to bind to the first. This dimerization may be the first step in the TCR multimerization noted in solution and on cell surfaces. The increased stability of the TCR complex with agonist ligands may allow full activation of the signaling cascade that is not obtained with the antagonists. It may also allow binding of coreceptors (CD4 or CD8) to further stabilize the complex.

SELECTION OF VARIABLE REGIONS OF THE -CHAIN OF THE TCR

Individual variable regions of the -chain (V) of the TCR are expressed mainly in either CD4 or CD8 subsets of peripheral T cells. This situation could be due either to preferential recognition of the MHC class I or class II molecules or to interaction between the V region and the CD4 or CD8 coreceptor. We previously used mutants of an -chain transgene to show that the selection specificity for MHC class I or class II molecules can be determined by a single amino acid residue in either complementarity-determining region 1 or complementarity-determining region 2. The position of these residues suggests that they interact with MHC molecules and makes it highly unlikely that they interact with coreceptors.

We have sequenced the expressed members of the V3 and V11 families from mice hemizygous for the TCR -chain. These mice are the progeny of a cross between normal mice and -chain knockout mice and can therefore only make a single species of -chain per cell. With these mice, we can avoid the problem of lack of allelic exclusion in the expression of -chain mRNA. We found that individual members of each family showed preferential expression in CD4 or CD8 cells. The position of sequence differences between the V regions strengthens our conclusion that "skewed" expression of V regions is due to preferential recognition of MHC contact residues. The residues that we determined are important for recognition of class I or class II MHC molecules are also important contact points between the TCR and the MHC in a complex composed of a TCR and a class I MHC molecule with bound peptide (as shown by I.A. Wilson and colleagues, Department of Molecular Biology). We are preparing mutant V3 soluble TCRs to determine the changes in affinity for a defined class I MHC--peptide complex that are produced by alterations in these single amino acid residues.

POLYMORPHISM IN THE CD4/CD8 RATIO AND THE TCR -CHAIN LOCUS

In collaboration with D. Lo, Department of Immunology, we have investigated the effects of V polymorphism on the T-cell repertoire. It has long been known that the CD4/CD8 ratio in mice (and, indeed, in humans) is polymorphic. Genetic segregation analysis indicated that this polymorphism maps closely to the locus of the TCR -chain. This finding was confirmed by using 2 independent pairs of congenic mouse strains to compare the Tcra^b haplotype with the Tcra^a and Tcra^c haplotypes. The results suggest that the V polymorphisms that determine a preference for MHC restriction also have an effect on the relative size of the CD4 and CD8 T-cell pools.

ALLELIC EXCLUSION OF THE TCR -CHAIN

Allelic exclusion of the TCR -chain is poor. Many mature T cells express 2 mRNAs capable of encoding -chain proteins, and many express 2 proteins. However, expression of 2 -chains on the cell surface is quite rare. We showed that functional allelic exclusion is attained in the thymus at the time that the cell-surface expression of TCRs is upregulated, after the start of positive selection. We have now shown that this exclusion is posttranslationally regulated. It is not caused by downregulation of one of the -chains at either the mRNA or the protein level. This finding suggests that phenotypic allelic exclusion depends on competition by the -chain proteins for pairing with the ß-chain. The limiting factor may be the CD3 -chain that stabilizes the TCR ß CD3-complexes. We are using -chain knockout and transgenic mice to test this possibility.

RECOGNITION AND EXPRESSION OF SUPERANTIGEN

The superantigens are a functionally defined group of proteins that activate a large percentage of T cells. These proteins recognize one of the relatively low number of variable regions of the ß-chain (Vß) of the TCR, rather than the almost infinite number of regular TCR combining sites. Superantigens are presented to the TCR after binding to MHC class II molecules. Different superantigens recognize different Vß elements, causing most or all of the cells bearing that particular Vß to be activated and then to die, with diverse pathologic effects.

Many different superantigens have been described, including those made by bacteria (e.g., staphylococci), mycoplasma, and various viruses such as mouse mammary tumor virus. We have expressed fragments of the superantigen of mouse mammary tumor virus in yeast. Some of these recombinant proteins activate T cells in vitro. We are trying to measure the affinity of these superantigens for MHC class II molecules and for TCRs. The results will enable us to determine which sites on the superantigen are involved in the interaction with its ligands.

We have prepared a series of monoclonal antibodies against recombinant mouse mammary tumor virus superantigen. Studies with these antibodies showed that the superantigen is expressed on the cell surface of B cells that express MHC class II molecules, as expected, and also on some CD8 (but not CD4) T cells. This latter finding is somewhat surprising, because the T cells do not express MHC class II molecules, and expression of these molecules is required for presentation of the viral superantigen to TCRs for activation of T cells. However, the results do fit with the finding several years ago of our colleagues S. Webb and J. Sprent that CD8 T cells strongly induce T-cell responses to viral superantigens in vivo. We are determining the detailed expression patterns and the biochemical characteristics of these molecules.

PUBLICATIONS

Alam, S.M., Gascoigne, N.R.J. Posttranslational regulation of TCR V allelic exclusion during T-cell differentiation. J. Immunol. 160:3883, 1998.

Gascoigne, N.R.J. Natural killer cells: Influence of the home environment. Curr. Biol. 7:R624, 1997.

Lo, D., Aftahi, N., Reilly, C., Neal, H., Sim, B.-C., Gascoigne, N.R.J., Kono, D., Wu, A., Schulman, S., Scott, B. Mapping genes regulating lymphocyte function: Correlations with autoimmunity? In: New Directions in Autoimmunity. Vol. 1. Genetics in Autoimmunity. Theofilopoulos, A.N. (Ed.). Karger, New York, in press.

Sim, B.-C., Aftahi, N., Reilly, C., Bogen, B., Schwartz, R.H., Gascoigne, N.R.J., Lo, D. Thymic skewing of the CD4/CD8 ratio maps with the T cell receptor -chain locus. Curr. Biol. 8:701, 1998.

Sim, B.-C., Lo, D., Gascoigne, N.R.J. Preferential expression of TCR V regions in CD4/CD8 subsets: Class discrimination or co-receptor recognition? Immunol. Today 19:276, 1998.

Sim, B.-C., Travers, P.J., Gascoigne, N.R.J. V3.2 selection in MHC class I mutant mice: Evidence for an alternate orientation of TCR-MHC class I interaction. J. Immunol. 159:3322, 1997.

Sim, B.-C., Wung, J.L., Gascoigne, N.R.J. Polymorphism within a TCRAV family influences the repertoire through class I/II restriction. J. Immunol. 160:1204, 1998.