Search

To determine the mechanism and structural consequences of peptide binding to class I molecules, we have studied the Ld molecule of the mouse. Previous studies have shown that a significant proportion of surface and intracellular Ld molecules can be detected in an alternative conformation designated Ldalt. Ldalt molecules are non-ligand associated and show weak if any beta 2-microglobulin (beta 2m) association. We report here that Ld molecules have a relatively rapid surface turnover compared with other class I molecules and that exogenous peptide dramatically prolongs Ld surface half-life. By contrast, Ldalt molecules are stably expressed on the surface and their half-life is unaffected by exogenous peptide. To study the surface interaction of peptide with Ld, live cells were incubated with iodinated peptides and Ld molecules were precipitated from cells precoated with monoclonal antibody before lysis. Using this assay, peptide binding to surface Ld molecules was found not to depend upon exchange with exogenous beta 2m, but did correlate with the level of beta 2m association. To study the intracellular interaction of peptide with Ld, cell lysates were used. In cell lysates, peptide was found to convert Ldalt molecules to properly folded Ld. This peptide-induced folding was almost complete at earlier but not later time points in pulse-chase analyses. Furthermore, conversion of Ldalt to Ld was found to affect almost exclusively immature (Endo Hs) class I molecules. Thus intrinsic properties of immature Ldalt molecules or their associated chaperonins are maintained in cell lysates that allow them to undergo de novo folding in vitro. These combined results demonstrate that immature Ldalt molecules are precursors awaiting constituents such as peptide and beta 2m that influence folding, whereas surface Ldalt molecules appear refractory to association with peptide, beta 2m, and consequent folding.

Two phenomena appear to distinguish the D region class I genes from those in the K region in the murine MHC: (a) haplotype disparity in the number of expressed D region class I molecules has been observed; and (b) clines of closely related D region class I molecules among and within mice of different H-2 haplotypes can be defined. Both of these observations have been based on serological and peptide mapping analyses of these molecules. Recent reports using molecular biological approaches have corroborated these findings. Since the mouse strain B10.AKM expresses multiple D region class I antigens, all of which are closely related to the prototypic Ld molecule, we investigated the Dq region of B10.AKM using molecular approaches. Three D region class I genes were isolated from genomic B10.AKM bacteriophage and cosmid libraries. Based on alignment of those genes with the BALB/c D region class I genes by analogous restriction endonuclease sites and by hybridization of one of those genes with a D4d gene-derived oligonucleotide probe, we have designated these genes as Dq, Lq, and D4q. As determined by DNA-mediated gene transfer to mouse L cells followed by serological analyses, the Dq and Lq genes encode previously characterized Dq region class I antigens. The nucleic acid sequence comparisons of the Dq and Lq genes demonstrated a higher level of homology with the Ld and Db genes than with other D region class I genes. In addition, CTL stimulated with a Dq, Lq, or Ld gene transfectant showed strong crossreactions with the other transfectants as targets, suggesting that the products of these genes are also functionally related. Thus, these studies suggest that the L molecule represents a prototypic structure shared by several D region gene products, and furthermore, the duplication of an Ld-like progenitor gene resulted in two Dq region class I genes, Dq and Lq. Unexpectedly, the sequences determined for the Dq and Lq genes are nearly identical to the sequences of two genes, A166 and A149, respectively, which were reported to encode the tumor-specific antigens; these novel class I genes were isolated from an H-2k fibrosarcoma, 1591. This raises the distinct possibility that these purported tumor-specific class I genes were introduced into this tumor by contamination.

To resolve issues regarding the evolution of D region class I MHC genes and their relationship to other class I-encoding regions of the mouse, as well as man, we characterized the class I genes from the Dq region of the B10.AKM mouse strain. The Dq region was selected because it was known to express multiple gene products, yet two of the products previously characterized have structural features in common with the Ld molecule. Since DNA hybridization data defined similarities between the Dd and Dq regions, we used low-copy genomic or oligonucleotide probes derived from the Dd region of BALB/c (H-2d) to screen a B10.AKM cosmid library. Cosmid clones containing Dq, D2q, D3q, D4q, Lq, and Q1q genes have been isolated and aligned with the corresponding genes of the BALB/c MHC, thus demonstrating a similar gene organization. The two classical transplantation genes, Dq and Lq were found to be strikingly similar to each other such that exons 1-3 of Dq and Lq, are approximately 97% homologous, and exons 4-8 are identical. Furthermore, the implied amino acid sequences of both Lq and Dq molecules show considerable homology to Ld, particularly in regions presumed to be involved in ligand binding. These comparisons suggest not only that the Dq and Lq genes arose from the duplication of an Ld-like progenitor, but also that there is a selective advantage for the maintenance of an Ld-like structure. In addition, the 5' portion of the D4q gene was sequenced and found to have a 13-bp deletion and a 4-bp insertion within the alpha 2 exon. These result in a frame shift that creates a premature termination codon and potential polyadenylation site, respectively. Thus, D4q does not encode a typical class I molecule. Sequence comparisons suggest that the D4q gene did not arise from a duplication event involving an Ld-like gene such as Dq and Lq. Interestingly, the D4q molecule, if produced, would have amino acid residues in common with K and/or Q molecules that differ from those observed in D/L molecules. These findings, in conjunction with hybridization data, provide evidence that the D2, D3, and D4 genes were derived from Q genes by an unequal crossover event. Additional hybridization data using low-copy D region probes suggest that several different D region gene organizations exist among mice of different haplotypes. These and other recent molecular studies provide multiple examples of expansion and contraction of the class I genes in the D region.(ABSTRACT TRUNCATED AT 400 WORDS)

To better understand the biological implications of the association of ligand with major histocompatibility complex class I molecules, we have studied the Ld molecule of the mouse. The culturing of various nonselected cell lines with three different known Ld peptide ligands resulted in a two- to fourfold specific increase in surface Ld expression as detected by 10 of 11 different monoclonal antibodies (mAbs) recognizing Ld epitopes. These findings suggest that Ld molecules are not saturated with endogenous peptide ligands and thus have accessible binding sites. Exploiting this feature of Ld we demonstrate that the physical association of Ld with ligand is exquisitely specific, indicating that they function in determinant selection. In addition, a non-peptide-bound antigenic variant of Ld was specifically detected with an exceptional mAb designated 64-3-7. In comparison with other Ld molecules, 64-3-7+ Ld molecules are not peptide ligand inducible, are more susceptible to proteolysis, lack beta 2 microglobulin association, and display a slower rate of oligosaccharide maturation. In spite of their deficiencies, the non-ligand-associated 64-3-7 Ld molecules were detected on the surface of all cell types tested; however, they appear not to be recognized by alloreactive cytotoxic T lymphocytes.