This proposal represents a continuation of our studies examining the structure-function of class I molecules of the MHC. Three issues will be addressed. The first will determine the role of the alpha3 domain of class I in the positive and negative selection of the T cell receptor (TcR) repertoire. We will utilize a laboratory engineered class I molecule, LQ3, which is the Ld in the alpha1, alpha2, TM, and cytoplasmic domains but derives its alpha3 domain from Q7b. Our published and preliminary data demonstrates that LQ3 binds Ld restricted peptides and is recognized by CD8 independent but not CD8 dependent cytotoxic T lymphocytes (CTL). The expression of LQ3 in C3H transgenic (TG) animals does not allow for the generation of cytomegalovirus (MCMV) specific Ld-restricted CTL, nor does it permit complete negative selection of anti-Ld alloreactive CTL. We will generate TG mice that express a TcR for Ld-MCMV and mate these animals to existing C3H.Ld and C3H.LQ3 TG mice to follow the fate of such cells. The failure of CTL to recognize LQ3 is presumably due to the inability of CD8 to interact with its ligand in the Q7 alpha3 domain. We will use a variety of CD8 <-> class I binding assays to characterize this defect. The second issue will address the relationship between low affinity (1o) and high affinity (2o) receptor alloreactive CTL. We will utilize the TcRs from a representative clone of each of these CTL to generate TcR TG mice. This will allow us to characterize the precursor-product relationship between such clones with regard to receptor avidity, CD8 expression, lymphokine secretion, and dependence of CD4 T helper activity. The third issue will focus on Qa-2 as an antigen presenting molecule. We will determine the TcR V gene usage of CTL specific for Q6 and Q7, the 2 major Qa-2 encoded antigens. We will determine the sequence of the Q6 gene, which is the major target molecule for Qa-2 specific CTL and establish whether these Qa molecules present a limited array of peptides.