The objectives of this project were to define and develop the methodologies for the in vitro immunization of human B lymphocytes in order to make it possible generate human monoclonal antibodies of desired specificity and affinity for antigen.
Various sources of B lymphocytes such as tonsils, peripheral blood and regional lymph nodes have been tested and it appears that the most general source, peripheral blood, can be used for most applications (Lund and Copenhagen). The in vitro immunization system has been adapted for the production of human monoclonal antibodies against a novel class of T cell independent antigens, low molecular weight glycolipids such as gangliosides, by incorporation into liposomes with or without a synthetic T helper epitope derived from tetanus toxin (amino acids 830-843). Using this system, it has been possible to produce highly specific IgM antibodies against ganglioside antigens such as GM3, a group of antigens which might have important therapeutic applications in cancer (Copenhagen). One of the laboratories (Heraklion) has designed and synthesized a number of antigens and heterotope antigens using linear solid-phase peptide synthesis. In addition, a new class of antigens, lipopeptides, have been synthesized. As a T cell epitope, the tetanus toxin peptide (830-843) has been selected by virtue of its broad binding to various HLA class II alleles. The main aim for one of the involved groups (Edinburgh) has been to establish an explant culture system for human lymphoid tissue, which could be used for in vitro immunization. Procedures for the preparation, storing and culturing tissue explants have therefore been compared with cell suspensions obtained from the same lymphoid tissue and shows important differences between the kinetics and magnitude of Ig-production. It has been studied which type of helper-T cells are supportive during the in vitro B cell response to antigens (Lund). These T cells, (i.e. CD4(+)CD45RO(+) T cells), have been used together with heterotope peptides, i.e. immunogens containing both a B and a T cell epitope. This has made it possible to design an in vitro immunization protocol that produced very specific human antibodies. By further manipulating the T cell compartment of this protocol, it has recently been possible to demonstrate an antigen-specific isotype-switched IgG response in vitro. This in vitro immunization format has been shown by means of a novel phage selection approach (see below) to result in an apparently affinity-maturated IgG immune response in vitro (Lund). The focus of one of the groups (Torino) has been the study of cell surface receptors involved in B lymphopoiesis and has concentrated on two molecules, CD44 and CD38. The results of this work point to an important role of CD38 in signalling in human B cells. Work in another group (Birmingham) has focused on the identification of the accessory molecules and cytokines regulating the growth and differentiation of germinal center (GC) B cells from tonsils. Proliferation of resting B lymphocytes can be maintained for many weeks in vitro in the absence of antigen and T cells by triggering the CD40 molecule with an anti- CD40 mouse monoclonal antibody in the presence of fibroblast L cells and the T cell cytokine IL-4. Under these conditions simultaneous binding of the monoclonal antibodies to CD40 on the B cell and to transfected Fcgamma receptors (FcgammaRs) on L cells. This method of maintaining B cells in vitro, known as the CD40 culture system, has been adapted in Birmingham and Lund to identify the cytokines and accessory molecules which are important for the growth and differentiation of B cells from germinal centres. We have now obtained not only specific IgG antibodies with high efficiency, but one of these systems has generated affinity maturation.
MAJOR SCIENTIFIC BREAKTHROUGHS:
A final goal of this project was to achieve an isotype- switch of the antigen specific IgM producing B cells generated during the primary in vitro immunization. To achieve this, we investigated the molecular and cellular requirements for antigen-specific isotype switch of human B cells by mimicking germinal centre events (Lund). Peripheral blood mononuclear cells from healthy seronegative blood donors were first primary immunized in vitro using the heterotope which generated specific IgM secreting B cells. We used the apex of the V3 loop of gp120 as the B cell epitope linked to a promiscuous T helper epitope from tetanus toxin. In parallel, CD4(+) helper cell clones specific for the T epitope of the immunogen were established. In a secondary in vitro stimulation period, we co- cultured the antigen-specific T and B cells on CD32-transfected fibroblasts, together with an anti-CD40 monoclonal antibody. This resulted in isotype switching and human antigen-specific IgG-secreting B cells were detected. Antigen-specific human B cells derived from this primary and secondary in vitro immunization were subsequently subjected to somatic cell hybridization and hybridomas secreting human anti-V3 IgG monoclonal antibodies were isolated. The third goal was to induce an antigen-specific, affinity maturated IgG response in vitro. To achieve this, we simulated germinal center events in vitro by introducing primary immunized human B cells to rescue and proliferation signals together with antigen-specific activated human T cells. We used the apex of the V3 loop of gp120 as a B cell epitope linked to a promiscuous T helper epitope from tetanus toxin (Heraklion). In parallel, CD4(+) T helper cell lines specific for the T epitope of the immunogen were stimulated. The results were evaluated with antibody library and phage display technologies. The increase in affinity was close to a 100-fold after a 12-day in vitro cultivation period which demonstrated a strong maturation of affinity during this immunization procedure (Lund).
A novel approach to the clonal selection of high-affinity human antibodies based on phage display has recently been developed in one of the participating laboratories (Lund). Filamentous bacteriophages lacking the phage minor coat protein 3 are non-infectuous because they are unable to attach to their host. This approach links antigenic recognition and phage-replication by producing a non-infectious phage library. The addition of a fusion protein between protein 3 and the antigen in this can the V3 loop of HIV-1 is then performed, and only the V3-specific phages will achieve the ability to infect, since they now carry the necessary infection-mediating protein 3.
Funding SchemeCSC - Cost-sharing contracts
220 07 Lund
B15 2TT Birmingham
EH8 9AG Edinburgh