Cesar Milstein

Production of monoclonal antibodies to group A erythrocytes, HLA and other human cell surface antigens-new tools for genetic analysis

Antibody-secreting hybrid cells have been derived from a fusion between mouse myeloma cells and spleen cells from a mouse immunized with membrane from human tonsil lymphocyte preparations. Hybrids secreting antibodies to cell surface antigens were detected by assaying culture supernatants for antibody binding to human tonsil cells. Six different antibodies (called W6/1, /28, /32, /34, /45 and /46 were analyzed. These were either against antigens of wide tissue distribution (W6/32, /34, and /46) or mainly on erythrocytes (W6/1 and W6/28). One of the anti-erythrocyte antibodies (W6/1) detected a polymorphic antigen, since blood group A1 and A2 erythrocytes were labeled while B and O were not. Antibodies W6/34, /45 and /46 were all against antigens which were mapped to the short arm of chromosome 11 by segregation analysis of mouse-human hybrids. Immunoprecipitation studies suggest that W6/45 antigen may be a protein of 16,000 dalton, apparent molecular weight, while W6/34 and /46 antigens could not be detected by this technique. Antibody W6/32 is against a determinant common to most, if not all, of the 43,000 dalton molecular weight chains of HLA-A, B and C antigens. This was established by somatic cell genetic techniques and by immunoprecipitation analysis. Tonsil leucocytes bound 370,000 W6/32 antibody molecules per cell at saturation. The hybrid myelomas W6/32 and W6/34 have been cloned, and both secrete an IgG2 antibody. W6/32 cells were grown in mice, and the serum of the tumor-bearing animals contained greater than 10 mg/ml of monoclonal antibody. The experiments established the usefulness of the bybrid myeloma technique in preparing monospecific antibodies against human cell surface antigens. In particular, this study highlights the possibilities not only of obtaining reagents for somatic cell genetics, but also of obtaining mouse antibodies detecting human antigenic polymorphisms.

Analysis of cell surfaces by xenogeneic myeloma-hybrid antibodies: differentiation antigens of rat lymphocytes.

Abstract A new approach to study differentiation antigens by means of monoclonal xenogeneic antibodies produced by myeloma-hybrid lines in culture is described. Spleen cells from mice immunized with rat thymocyte membranes were fused with a mouse myeloma. Hybrids were selected and screened for the production of antibodies to rat thymocytes by a binding assay. A number of such antibodies was detected, and five specificities (W34, /6, /13, /15 and /25) were further investigated. Two of the hybrid myelomas were cloned, and the monoclonal antibodies were shown to contain heavy chains of molecular weight 50,000 daltons (W313) and about 70,000 daltons (W34), respectively. All antibodies were shown to detect different, previously undefined antigens on different subpopulations of lymphoid cells. The different labeling patterns on thymocytes, thoracic duct lymphocytes and bone marrow cells were investigated by the fluorescence-activated cell sorter. Antibodies W36, /13, /15 and /25 labeled most thymocytes, while W34 labeled very few. Among thoracic duct lymphocytes, W313 labeled most, if not all, of the T cells, and W325 labeled a large fraction of them; W34, on the other hand, labeled mainly B cells. In bone marrow, W313 and W315 labeled sub-populations which were largely independent of each other. Absorption studies showed that brain, but not kidney or liver, absorbed W313, while all these tissues were markedly less effective than lymphoid cells for absorption of W325. The number of antibody molecules per cell binding at saturation to thymocytes was: 5000 (W36); 17,000 (W315 and W325); 36,000 (W313). None appears, therefore, to detect major proteins in the membrane. The method is therefore extremely sensitive and allows identification down to minor membrane molecules and also of antigens on small subpopulations of a heterogeneous mixture of cells.

Mutation drift and repertoire shift in the maturation of the immune response.

The halltnark of the immune response is its specificity and the specificity is directly correlated with the affinity of the antigen-antibody interaction. The requirement for high affinity antibodies may be more important than specificity alone, since antibodies are designed to detect soluble antigens which are sometimes capable of inflicting great harm at very low concentrations (e.g. toxins). This may not be required by. or may even be a disadvantage to T-cell responses where the affinity for the ligand involves interactions of the T-cell receptor not only with antigen, but also with other molecules, e.g. those involved in MHC restriction (Yague et al. 1985. Dembic et al. 1986). T cells therefore may not have developed the equivalent of the elaborate mechanism which is the object of this paper. During the course of an antigen-specific immune response, the affinity of the serum increases with time, a phenomenon commonly referred to as maturation of the response (Jerne 1951, Siskind & Benaceraff 1969). Such a maturation results from specific alterations of the structure of the antibody molecules (Steiner & Eisen 1967). What is the precise nature of these alterations, which are the root of the production of high affinity antibodies? There is no doubt that somatic mutation contributes to antibody diversity (Weigert et al. 1970. Bernard et al. 1978, Griffiths et al. 1984). There are many reasons to believe that a mechanism of hypermutation operates within restricted stretches of the DNA to further diversify the genes encoding the antibody molecules (Kim et al. 1981, Gearhart & Bogenhagen 1983). This mutational drift is, however, not the full extent of the change. Major changes in the antibody structures involved result from a shift in the antigen-specific B-cell repertoire over the course of the immune response. In the primary response the most frequent B-cell clones already expressing antibody molecules with a relatively

The Dynamic Nature of the Antibody Repertoire

Antibody production by an animal in response to an immunogen is dictated by the repertoire of B cells at the time of stimulation. The repertoire of B cells, however, is continuously changing. This change is brought about by the dynamic character of the repertoire, and by the antigenic experience of the animal. The naive repertoire is formed by genetic recombination events which are independent of antigenic stimulation. Since the size of the B-cell repertoire only allows the expression of a fraction of the genetic repertoire at any given time, only a random sample of the total potential is available. This constitutes the raw material that must be capable of providing initial recognition to any antigenic challenge to which animals are subjected. This raises the question of how an animal discriminates between self and non-self antigens, an aspect of the response we will not discuss here. We want to restrict ourselves to points in which our experience of the anti-oxazolone response gives informative data. We will therefore concentrate our attention on describing the way in which the antigenic experience influences the available B-cell repertoire. Our general ideas on the subject will be discussed in the context of the model presented in Fig. 1. We will argue that antigensensitive B cells from both the naive and the memory repertoire are modified through hypermutation and selection following each round of antigenic stimulation, so as to continuously refme the pool of memory B cells. One of the important conclusions we draw from otir experiments is that the antigetiic stimulation of memory cells leads to further processes of hypermutation and selection, functioning through a highly organized and purpose-developed procedure, probably operating in the germinal centers (see Fig. 1). We wish to argue that the development of such an elaborate procedure, so far unique to antibody-producing B cells, arose in response to the evolutionary advantage

Discriminating intrinsic and actigen-selected mutational hotspots in immunoglobulin V genes

Studies of the antibody hypermutation mechanism have revealed that it is not a random process but exhibits characteristic nucleotide substitution preferences. Here, Alexander Betz and colleagues show that these innate nucleotide substitution preferences can be used to examine databases of antigen-selected V gene sequences and thereby distinguish intrinsic from antigen-selected hotspots. This analysis reveals intrinsic mutational hotspots in both VH and VL genes, reflecting innate features of the hypermutation machinery which may give clues to the enzymatic mechanism.

Developmental regulation of IgM secretion: The role of the carboxy-terminal cysteine

B lymphocytes do not secrete IgM, and plasma cells only secrete IgM polymers. Here we show that both events are attributable to the tailpiece found at the carboxyl terminus of mus chains, and we specifically implicate Cys-575. Thus, if Cys-575 was mutated, IgM was secreted by B cells. Similarly, a mutant IgG containing a mus tailpiece became largely retained within the cell; secretion was restored upon mutation of the tailpiece cysteine. Removal of Cys-575 also allowed hypersecretion of monomeric IgM by plasmacytoma cells. Following further removal of Cmu1, heavy chains were secreted in the absence of light chains. Thus, in B and plasma cells, Cys-575 is involved both in the polymerization of IgM and in intracellular retention of unpolymerized intermediates.

Human CD1d-glycolipid tetramers generated by in vitro oxidative refolding chromatography.

CD1 molecules are specialized in presenting lipids to T lymphocytes, but identification and isolation of CD1-restricted lipidspecific T cells has been hampered by the lack of reliable and sensitive techniques. We here report the construction of CD1d–glycolipid tetramers from fully denatured human CD1d molecules by using the technique of oxidative refolding chromatography. We demonstrate that chaperone- and foldase-assisted refolding of denatured CD1d molecules and β2-microglobulin in the presence of synthetic lipids is a rapid method for the generation of functional and specific CD1d tetramers, which unlike previously published protocols ensures isolation of CD1d tetramers loaded with a single lipid species. The use of human CD1d–α-galactosylceramide tetramers for ex vivo staining of peripheral blood lymphocytes and intrahepatic T cells from patients with viral liver cirrhosis allowed for the first time simultaneous analysis of frequency and specificity of natural killer T cells in human clinical samples. Application of this protocol to other members of the CD1 family will provide powerful tools to investigate lipid-specific T cell immune responses in health and in disease.

Mouse CD1 is distinct from and co-exists with TL in the same thymus.

Human CD1 antigens have a similar tissue distribution and overall structure to (mouse) TL. However recent data from human CD1 suggest that the mouse homologue is not TL. Since no human TL has been conclusively demonstrated, we have analysed the murine CD1 genes. Two closely linked genes are found in a tail to tail orientation and the limited polymorphism found shows that, as in humans, the CD1 genes are not linked to the MHC. Both genes are found to be equally transcribed in the thymus, but differentially in other cell types. The expression in liver, especially, does not parallel CD1 in humans. This demonstrates conclusively that CD1 and TL are distinct and can co‐exist in the same thymus. It is paradoxical that despite the structural similarity between mouse and human CD1, the tissue distribution of human CD1 is closer to TL. The possibility of a functional convergence between MHC molecules and CD1 is discussed.

AID-GFP chimeric protein increases hypermutation of Ig genes with no evidence of nuclear localization

Somatic hypermutation generates variants of antibody genes and underpins the affinity maturation of antibodies. It is restricted to the V-gene segments, and although it decays exponentially toward the 3′end, it includes recognizable hot spots. Although the detailed mechanism of hypermutation remains elusive, the process may take place in two separate stages, preferentially targeting G/Cs in the first and A/Ts in the second stage. It seems that MSH2 is involved in the second stage, and that activation induced deaminase (AID) is implicated in the control of hypermutation. The constitutively hypermutating cell line Ramos expresses AID, and we have prepared transfectants that express a chimeric AID-green fluorescent protein. The fluorescence is strongly detected in the cytoplasm but not in the nucleus. Yet, the chimeric protein increases the hypermutation rate either directly or, more likely, indirectly, by favoring the transport of AID into the nucleus. Thus, in Ramos, AID seems to be rate limiting. Unexpectedly, the proportion of deletions also is increased. The increase in mutation rate detected by a fast cytofluorimetric method based on the accumulation of sIgM-loss mutants correlates with the increase measured by mutations defined by sequence analysis. The higher mutation rate is largely explained by the higher proportion of mutated clones, indicating that AID controls the number of cells that undergo hypermutation but not the number of mutations that are incorporated in each mutation round.

Switch junction sequences in PMS2-deficient mice reveal a microhomology-mediated mechanism of Ig class switch recombination

Isotype switching involves a region-specific, nonhomologous recombinational deletion that has been suggested to occur by nonhomologous joining of broken DNA ends. Here, we find increased donor/acceptor homology at switch junctions from PMS2-deficient mice and propose that class switching can occur by microhomology-mediated end-joining. Interestingly, although isotype switching and somatic hypermutation show many parallels, we confirm that PMS2 deficiency has no major effect on the pattern of nucleotide substitutions generated during somatic hypermutation. This finding is in contrast to MSH2 deficiency. With MSH2, the altered pattern of switch recombination and hypermutation suggests parallels in the mechanics of the two processes, whereas the fact that PMS2 deficiency affects only switch recombination may reflect differences in the pathways of break resolution.