The binding of a T-cell antigen receptor (TCR) to peptide antigen presented by major histocompatibility antigens (pMHC) on antigen-presenting cells (APCs) is a central event in adaptive immune responses. The mechanism by which TCR-pMHC ligation initiates signalling, a process termed TCR triggering, remains controversial. It has been proposed that TCR triggering is promoted by segregation at the T cell-APC interface of cell-surface molecules with small ectodomains (such as TCR-pMHC and accessory receptors) from molecules with large ectodomains (such as the receptor protein tyrosine phosphatases CD45 and CD148). Here we show that increasing the dimensions of the TCR-pMHC interaction by elongating the pMHC ectodomain greatly reduces TCR triggering without affecting TCR-pMHC ligation. A similar dependence on receptor-ligand complex dimensions was observed with artificial TCR-ligand systems that span the same dimensions as the TCR-pMHC complex. Interfaces between T cells and APCs expressing elongated pMHC showed an increased intermembrane separation distance and less depletion of CD45. These results show the importance of the small size of the TCR-pMHC complex and support a role for size-based segregation of cell-surface molecules in TCR triggering.

Vaccinia infection interferes with the presentation of influenza Haemagglutinin (HA) and Nucleoprotein (NP) to class I-restricted CTL. The inhibitory effect is selective for certain epitopes, and is more profound during the late phase of infection. For influenza A/NT/60/68 NP, the block is present during both early and late phases of infection, and is selective for the COOH-terminal epitope defined by peptide 366-379, having no detectable effect on the presentation of the NH2-terminal epitope 50-63. The presentation of HA is inhibited only during the late phase of vaccinia infection. For both proteins, presentation is partially (NP) or completely (HA) restored by expression of rapidly degraded protein fragments in the vaccinia infected target cell. For HA, deletion of the NH2-terminal signal sequence completely overcomes the block. For NP, either a large NH2-terminal deletion or the construction of a rapidly degraded ubiquitin-NP fusion protein partially restores presentation. These results illustrate the relationship between degradation of viral proteins in the cytoplasm of an infected cell and recognition of epitopes at the cell surface by class I-restricted T cells.

The mRNA sequence of the human intrinsic clotting factor IX (Christmas factor) has been completed and is 2802 residues long, including a 29 residue long 5′ non‐coding and a 1390 residue long 3′ non‐coding region, but excluding the poly(A) tail. The factor IX gene is approximately 34 kb long and we define, by the sequencing of 5280 residues, the presumed promoter region, all eight exons, and some intron and flanking sequence. Introns account for 92% of the gene length and the longest is estimated to be 10 100 residues. Exons conform roughly to previously designated protein regions, but the catalytic region of the protein is coded by two separate exons. This differs from the arrangement in the other characterized serine protease genes which are further subdivided in this region.

A surprising feature of most cytotoxic T lymphocytes (CTL) responding to influenza infection is that they recognize the unglycosylated (non-transmembrane) proteins of the virus, including the nucleoprotein. Recognition of cells that express nucleoprotein by CTL does not depend on a definite signal sequence within the protein, and the epitopes recognized can be defined with short synthetic peptides in vitro. Haemagglutinin (HA), the major transmembrane protein of the virus, is recognized by a minor population of CTL from infected mice. We have deleted the sequence coding for the N-terminal signal peptide from a complementary DNA encoding HA of the H1 subtype. The signal-deleted HA is detected with antibodies as a short-lived, unglycosylated, intracellular protein. However, CTL raised to the complete molecule recognize cells expressing the signal-deleted HA and vice versa. These results cast doubts on the assumption that CTL recognize the HA molecule only after its insertion into the plasma membrane.