Generation of diverse high-affinity human monoclonal antibodies by repertoire cloning.

Abstract

Combinatorial libraries of antibody heavy and light chains derived from the peripheral blood lymphocytes of an individual immunized with tetanus toxoid have been expressed in Escherichia coli by using phage A vectors. Screening of the libraries allowed identification of a large number of human monoclonal Fab fragments specific for tetanus toxoid.Initial studies suggested considerable sequence diversity in these antibodies. The method should allow the generation of many human monoclonal antibodies of interest and the dissection of human humoral immune responses.Mouse monoclonal antibodies are readily generated by the fusion procedure of Kohler and Milstein (1). However, for therapeutic applications, human monoclonal antibodies are preferred. Despite extensive efforts, including production of heterohybridomas (2), Epstein-Barr virus immortalization of human B cells (3) and "humanization" of mouse antibodies (4), no general method comparable to the Kohler-Milstein approach has emerged for the generation of human monoclonal antibodies (reviewed in ref. 2). Previously, a procedure was described for the production of mouse monoclonal antibodies based on antigen selection from a combinatorial library ofthe mouse antibody repertoire cloned in phage A (5). The procedure has recently been successfully applied to the isolation of mouse monoclonal antibodies specific for influenza virus hemagglutinin (6). The extension to human antibodies is not trivial. Mice can be hyperimmunized with antigen and the spleen, a rich source of antibody-producing cells, can be removed. Humans can be immunized or are immune to many antigens but only peripheral blood, a poor source of antibody-producing cells, is readily available.Here we show that the repertoire cloning approach can be applied for the generation of human monoclonal antibodies from peripheral blood lymphocytes (PBLs) and we speculate that, given appropriate donor selection, this may effectively constitute a general route to such antibodies. MATERIALS AND METHODSImmunization and Lymphocyte RNA Preparation. A healthy volunteer who had received the most recent tetanus toxoid immunization 18 months before this experiment was given a booster injection of 0.5 mg of diphtheria/tetanus toxoid intramuscularly. Seven days later, he was subjected to leukapheresis, processing 5 liters of blood resulting in 2.1 x 109 lymphocytes after additional density-gradient separation (Ficoll-Paque; Pharmacia); 0.5 x 109 cells each were taken for three parallel experiments: (i) untreated PBLs, RNA was isolated immediately by a guanidinium isothiocyanate method (7); (ii) in vitro antigen stimulation, in which the cells were cultured in the presence oftetanus toxoid (10 ,ug/ml) for 3 days in RPMI 1640 medium with 10%o human AB serum (GIBCO), after which the RNA was isolated as described above; (iii) panned antigen reactive cells, lymphocytes were incubated for 45 min with 60 nM biotin-tetanus toxoid, washed twice, and then poured onto Petri dishes coated with streptavidin and blocked with bovine ...

Evidence for a bias toward intracellular antigens in the local humoral anti-tumor immune response of a colorectal cancer patient revealed by phage display

“…The V-genes of these anti-TT antibodies were sequenced and shown to originate from different germline segments of different V-gene families: VH-genes were derived from the VH1 and VH4 (RP) or VH1 only (IgG) family; light chain genes originated from the VI and VIII (RP) or V1 and VIII (IgG) subgroups. As has been observed previously, 27 light chain promiscuity was seen for two of the antibodies selected from the IgG repertoire, where the same VH segment was found in combination with two different VL-genes. The frequency of somatic mutation in the V-genes of these selected antibodies (as determined by alignment of the sequences to their closest germline match) was high: overall, 21.8 mutations were found in the selected VH-genes, of which 15 caused an amino acid change in the expressed protein.…”

Human recombinant Puumala virus antibodies: cross-reaction with other hantaviruses and use in diagnostics.

“…The donor was chosen because she had a high titre of serum IgG reactive with the N and envelope proteins of PUU, and monoclonal antibodies (MAbs) to PUU had previously been isolated from her spleen cells using a cellular approach (Lundkvist et al, 1993). RNA was isolated from the spleen lymphocyte fraction and a cDNA library of expressed antibody genes was prepared by PCR amplification of IgG1 heavy chain Fd regions and whole λ and κ light chains in the phage-display vector pComb3H, in essence as described previously Persson et al, 1991).…”

Examination of a role for idiotypy in the disease remission of a long-term survivor of adult T cell leukemia treated with anti-Tac antibody

“…Additionally, cc␣Id showed no binding to plates coated with purified IL2 by ELISA. The diversity of the six libraries employed is more than sufficient to recover active antibody responses, 21,28,45,46 but it is uncertain that it could detect clones that are much more rare. An ongoing network-based antibody response is ruled out by the failure to detect IL2-reactive Ab2 or sTac-reactive Ab3 antibodies in the blood or to retrieve IL2-reactive or sTacreactive clones by phage display technology.…”