Monoclonal antibody detection of carbohydrate-defined and protein-defined H-2Kk antigens

Identification of molecules detected by three different anti-H-2Kk monoclonal antibodies

Three different monoclonal antibodies specific for murine Class I H-2Kk-encoded determinants have been found to bind to molecules other than the classical 45 Kd Class I glycoprotein. These results were obtained after electrophoresis of a detergent lysate of spleen cells, and extraction of molecules present in multiple gel fractions. The antigenic activity in each gel fraction was assessed after removal of detergent by capacity to inhibit the binding of each of these antibodies to Ig-capped spleen cells in a rosetting assay. Only the H100-27R9 antibody was inhibited by an extract representing 45-50 Kd proteins resembling Class I molecules. This antibody also bound material extracted from a higher 65-70 Kd fraction of the gel. The binding of the 11-4.1 and H100-30R3 antibodies was inhibited by unique molecules in the molecular weight range of less than 20 Kd. The inhibitory material was not sensitive to pronase but was sensitive to glycosidases. This material could be absorbed by each of the H100-30R3 and 11-4.1 antibodies but not by H100-27R9. All data are consistent with unique, low molecular weight carriers of carbohydrate-defined epitopes which can bind monoclonal antibodies having specificity for H-2K-encoded gene products. It is predicted that these are glycolipids, but it is not yet known whether they are cell-surface or cytoplasmic molecules.

Dependence of mouse thymocyte-erythrocyte rosette formation on complete identity at class-I-MHC

Mouse thymocytes and erythrocytes form rosettes when incubated together at 4 degrees C. The frequency is much higher when the thymocytes and erythrocytes are MHC-identical. If the indolizidine alkaloid swainsonine (SW) is present during rosette formation at concentrations of 1 microgram/ml (5.7 microM) or greater, rosette formation between MHC-identical pairs is inhibited to levels comparable to those observed for MHC-different pairs; rosette formation by MHC-different pairs is not affected. This was confirmed by examining 17 different MHC-identical combinations (9 completely syngeneic and 8 differing in non-MHC genes) and 13 MHC-different combinations (3 of these identical everywhere except at MHC). A SW-inhibitable component of rosette formation was observed only when thymocyte and erythrocyte were completely identical at MHC. Thus F1-parent pairs behaved as if allogeneic, although both F1-F1 and parent-parent had a SW-inhibitable rosetting component. Similarly, inbred strains only partially MHC-identical (B10.BR-B10.A, B10.D2-B10.A) behaved as if allogeneic. The SW-inhibitable component of rosetting could be partially but significantly blocked by including monoclonal antibodies against Thy-1, and anti-CD4 plus anti-CD8 (together but not separately); anti-class-I-MHC produced some inhibition of marginal significance. Monoclonal antibodies against class-I-MHC, LFA-1, and CD3 did not block. Pretreatment of erythrocytes with neuraminidase, greatly reduced the SW-inhibitable component of rosetting. The SW effect would appear to be due to a direct interaction of SW with a cell surface structure involved in syngeneic rosette formation rather than the known ability of SW to block the processing of N-linked sugar structures. The results are consistent with cell surface lectins and cell surface sugars playing a role in rosette formation.

Monoclonal antibodies which identify carbohydrate-defined MHC class I epitopes

Eleven different monoclonal antibodies specific for H-2K- and H-2D-encoded Class I molecules have been screened to determine Class I epitopes dependent on both carbohydrate and protein structures. Monoclonal antibodies have been identified which bind to carbohydrate-defined antigens encoded by both the H-2K and H-2D gene regions. Sensitivity to glycosidases versus pronase has been used to classify antigens both expressed as cell surface molecules and when prepared as detergent solubilized antigen. Several simple sugars have also been found to act as inhibitors of antibodies which bind to carbohydrate-defined sites. The genetic control of carbohydrate antigen expression by H-2K- and H-2D-linked genes has been verified since a specific antibody does not bind to H-2Kb or H-2Db molecules encoded by several mutant strains of mice containing single amino acid substitutions in their protein product. All of these data are consistent with Class I antigenic structures being encoded in carbohydrate as well as protein moieties.

Identification of class I H-2Db molecules primarily expressed by B lymphocytes in murine spleen

Evidence is presented which supports the phenomenon of heterogeneity amongst H-2Db-encoded Class I molecules. Two monoclonal antibodies (MoAb) called H141-31 and B22-249 were used in these studies. Both bind to the 'private' H-2.2 site of H-2Db-encoded molecules, but the binding of B22-249 is determined by carbohydrate moieties, whereas H141-31 appears to bind to a protein-defined epitope. Some H-2Db molecules, identified by the H141-31 MoAb, are primarily expressed on B lymphocytes and not T lymphocytes in spleen. The number of H-2Db molecules which bind H141-31 on B cells was also found to be three- to four-fold less than the number which bound the B22-249 MoAb. B cells of two mutant strains of mice, B6-C.H-2bm13 and B6-C.H-2bm14 which harbour very few nucleotide changes in the H-2Db gene, also show marked reduction in the binding of both antibodies. This suggests that a single common gene encodes both target molecules and that post-translational modifications such as differential glycosylation may account for heterogeneity amongst H-2Db molecules. This would explain the presence of the different H-2Db molecules defined here. It follows that differences in glycosylation evidently occur both within the B cell population, since H141-31 binds to only a subset of H-2Db molecules on B cells, and between T and B lymphocytes, since resting T cells do not bind H141-31 MoAb.

Detection of class I MHC antigens on glycosidase-treated cells

Several Class I molecules encoded by the murine H-2K gene have been found to express epitopes with different sensitivities to glycosidases. This result was found to be variably represented by different assays used to measure the binding of monoclonal antibodies (MoAb) specific for Class I H-2Kk molecules.

A trans-acting major histocompatibility complex-linked gene whose alleles determine gain and loss changes in the antigenic structure of a classical class I molecule

The RT1.A locus of the rat MHC encodes the H chain of the single classical class I molecule of this species. One of the alleles of this polymorphic locus, RT1.Aa, is present in several laboratory inbred, congenic, and MHC recombinant rat strains. Studies of the RT1.Aa class I molecule from a number of these strains as a target for CTL show that its antigenicity, both as an alloantigen and a restricting element, is subject to gain and loss alterations by the action of a gene mapping in the MHC to the right of RT1.A. This locus is apparently present in two allelic forms (one possibly a null allele) corresponding to the presence or absence of a dominant transacting modifier, and has been named class I modification, or cim. The antigenic change brought about by cim is scarcely detectable serologically but highly immunogenic for CTL. Biochemical investigations show that cim affects the post-translational modification of RT1.Aa.

A quantitative analysis of H-2 linked effects on hepatic ganglioside composition

We have conducted a quantitative analysis of the gangliosides extracted from brain, spleen, thymus, and liver tissue of 8-week-old male mice from H-2 congenic mouse strains on the B10 background, using high performance thin-layer chromatography (HPTLC). An analysis of variance of replicate samples of liver from strains B10, B10.A, and B10.G revealed that the time of sample and colony of origin were not sources of significant variance but that for N-glycolylated gangliosides GM2, GM1, and GD1a, the differences detected between strains were significant. Particularly important were the differences for GM1: the values of 0.0% for B10, 19.0% for B10.A, and 36.0% for B10.G were each significantly different from the others (P less than 0.0005). Further studies with liver tissue from B10/A H-2 recombinant strains also revealed three significantly different levels of GD1a: less than or equal to 4.0% [B10, B10.A (3R), B10.A (5R), B10.A (18R)], 11.0% (B10.A), and 33% [B10.A (1R), B10.A (2R), B10.A (4R)]. Our findings support prior studies which indicate that a gene linked to the H-2 complex affects hepatic GM2 galactosyltransferase activity. However, they also indicate that the current model, which classifies all strains as possessing either an allele for "high" enzyme activity or a single alternative allele for "low" enzyme activity, is probably oversimplified, since at least three levels of enzyme activity appear to exist as stable phenotypic markers. Moreover, the current model cannot readily account for the three different levels of GD1a observed with B10/A H-2 recombinants. Alternative models are proposed, including the novel suggestion that a distinct H-2 linked gene may affect hepatic GM1 sialyltransferase activity. These findings demonstrate that further study of H-2 linked genes affecting the activities of glycosyltransferases is indicated.

Amino acid sequence analysis of the H-2Kk alloantigen: complete sequence of residues 1-98 and partial sequence from 99 to 263

The H-2Kk molecule was purified by immunoprecipitation from the glycoprotein fraction of Nonidet P-40 extracts of RDM 4 mouse tumor cells. Cyanogen bromide cleavage of the major papain fragment yielded three peptides, the largest of which consisted of three disulfide-linked peptides which could be separated after reduction and alkylation. These peptides were readily aligned by their homology to similar fragments derived from other H-2 class I molecules. Amino acid sequence analyses of the two nondisulfide-linked peptides, peptide E (residues 1-52) and peptide D (53-98), yielded the following NH2-terminal sequence for the H-2Kk molecule: [sequence in text]. Comparison of this sequence with those of other H-2 class I molecules revealed that: (1) Lys-19, Val-55, Glu-56, Asn-63 and Ile-73 are unique to the H-2Kk molecule; and (2) H-2Kk shares 79-83% homology in this region with other mouse class I molecules. Partial NH2-terminal amino acid sequences are also reported for the three disulfide-linked peptides. Several discrepancies from previously reported partial sequences of the H-2Kk molecule were detected.

Alien antigens on tumour cells revisited

New findings on the detection of alien histocompatibility antigens (AHA) on tumour cells are reviewed. Three points are discussed, namely: (1) isolation and molecular characterization of AHA; (2) frequency of expression of AHA by tumour cells, and (3) mechanisms of expression of AHA. Mechanisms of expressions of AHA are particularly discussed in the light of the recent advances in the understanding of the molecular genetic organization of H-2 antigens. Gene conversion involving coding or non-coding sequences might be responsible for the appearance of unexpected H-2 determinants on cancer cells.

Epitopic analysis of detergent-solubilized HLA molecules by solid-phase radioimmunoassay

A solid-phase radioimmunoanalysis of plasma membrane molecules solubilized in the presence of detergent was developed. From a crude cell lysate, membrane molecules were specifically immobilized by solid-phase adsorbed monoclonal antibodies. Analysis of these molecules was easily performed with radiolabeled monoclonal antibodies of different specificities. This technique was used to analyze the HLA-DR and HLA-A,B,C molecules expressed by RAJI cells. It was possible (a) to determine which epitopes are expressed on the same molecules; (b) to define subsets of HLA molecules according to the epitopes which they express, and (c) to perform quantitative analysis of HLA molecules on a crude cell lysate.

Primary structure of the H-2Db alloantigen. II. Additional amino acid sequence information, localization of a third site of glycosylation and evidence for K and D region specific sequences

The complete amino acid sequence of the CNBr fragment comprising residues 229-284 of the murine major histocompatibility complex antigen H-2Db has been determined using radiochemical methodology. The sequence was determined by N-terminal sequence analysis of the intact CNBr fragment and by sequence determinations of peptides derived from this fragment by trypsin and staphylococcal V8 protease cleavage. In addition to the amino acid assignments for H-2Db, it was possible to assign the linkage position of the third N-linked glycosyl unit to the asparagine at residue 256. Additional amino acid sequence assignments have also been made for three other CNBr fragments that span residues 99-138, 139-228, and 308-331 of the H-2Db molecule. The total protein sequence information available (222 of 338 residues) agrees in every comparable position with the protein sequence derived from the cDNA clone (pH203) isolated by Reyes and co-workers (1982b), which strongly suggests that this clone encodes H-2Db. Combination of the protein sequence with that deduced from the cDNA clone provides the complete H-2Db protein sequence. Comparison of this sequence with other available protein sequence information for murine class I molecules has revealed protein sequences that may be unique to either K or D region molecules.