II. Gross Structure

The overall structure of the TCR is strikingly similar to that of the immunoglobulin Fab fragment, as predicted from the sequences and recombination events of TCR genes. The TCR is a heterodimer of one alpha and one beta polypeptide chain, joined by an inter-chain disulfide bond at the membrane-proximal, carboxy-terminal end of the molecule. The alpha and beta chains each contain clearly defined variable (V) and constant (C) domains. The molecule also contains four N-linked carbohydrate residues which extend away from the receptor into the solvent.

Three of the four TCR domains exhibit the typical immunoglobulin-type fold structure, with a "sandwich" of beta-sheets held together by an intra-chain disulfide bond. The secondary structure of the Ig-type fold is maintained by critical framework residues that are conserved throughout the immunoglobulin superfamily.

The Cα domain, however, exhibits a non-classical Ig-type fold. The participating cysteine residues are separated by a loop of only 49 amino acids, as opposed to the typical 70-80 residue spacing seen in immunoglobulins. This destroys the integrity of the outer beta sheet, with the remaining residues left to form two isolated strands broken by a mini-helix. As a result, the hydrophobic surface of the inner sheet is largely exposed to solvent. This may contribute to the observed instability of the alpha chain during TCR synthesis, and may indicate a special role of Cα in binding the CD3 co-receptor zeta chain through hydrophobic interactions.

Another unusual feature of the TCR is a 13-amino acid insertion in the beta chain (220-232β). This insertion forms a solvent-exposed loop extending from the elbow region of the molecule. The loop has a distinctive charge distribution, with three glutamate, one aspartate, and two lysine residues extending from the body of the molecule. The only contact made with the TCR domains is achieved by the lone hydrophobic residue (Trp225β) forming a weak interaction with the bridge between Vβ and Cβ. The lack of specific contacts to anchor the loop suggests some role other than maintaining the TCR domain structure, such as binding co-receptors or aiding in multimerization. Interestingly, this insertion is found only in the TCR beta and gamma chains.