A complete matched set of monoclonal antibodies consisting of the human isotypes IgM, IgG1 (allotype G1m(za), IgG2, IgG3 (allotypes G3m(b) and G3m(g)), IgG4, IgA2 and IgE was constructed with specificity for the hapten NP and its derivative NIP (5-iodo-4-hydroxy-3-nitrophenylacetyl) . The antibodies were affinity purified on hapten columns. The hapten, available as either a lipid soluble molecule NIP-kephalin, or as a protein modifier NIP-succinimide ester, was used to label cell surfaces. In this way the antibodies could be assayed for activation of effector mechanisms leading to cell killing.
The antibodies were tested for their abilities to mediate autologous complement-dependent lysis of human red blood cells labelled with the NIP-kephalin derivative and for antibody dependent cell-mediated cytotoxicity (ADCC) by activated human mononuclear cells of a human lymphoblastoid cell line coupled with NIP-succinimide ester. The results from these experiments are summarised in Figure 1 , shown as the results relative to the IgG1 isotype, and have been adapted from Bruggemann et al . The IgG1 antibody proved to be the most effective in both complement-dependent and cell-mediated cytotoxicity. The concentration of IgG1 antibody needed for 50% of maximal killing was approximately 0.1 µg/ml for complement lysis and for ADCC. The effector cells in ADCC were inhibited by a CD16 (Fc gamma RIII) monoclonal antibody and had the phenotype of killer-cells (K-cells). As expected, the IgM antibody proved to be good at complement activation but poor in ADCC. Surprisingly the two IgG3 antibodies were not quite as good as the IgG1 antibody in either assay. For complement, this was despite the fact that IgG3 was shown to be fixing many more molecules of C1q. Bindon et al  later went on to show that although human IgG1 bound less C1 than human IgG3 there was a much more efficient deposition of C4b on the cell surface which accounted for the more effective cell lysis by IgG1. None of the other isotypes IgG2, IgG4, IgA2 and IgE showed any significant functional activity in these assays.
The cell lines making this matched panel of antibodies have been widely distributed and are available from the European Collection of Animal Cell Cultures (Division of Biologics, PHLS Centre for Applied Microbiology and Research, Porton Down, Salisbury, SP4 0JG,UK), and purified antibody is commercially available. Other groups making use of this set of antibodies have extended the original observations. Two groups have investigated the effect of varying the antigen density as well as the epitope patchiness for complement lysis triggered by the NP chimaerics ,. The IgG1 antibody was most effective when the antigen concentration was higher, and whereas the IgG3 antibody was relatively better at lower concentrations. The IgG2 antibody gave good lysis at very high concentrations of antigen but the IgG4 antibody did not give lysis under any conditions . It was also shown that IgG1 and IgG3, along with IgM activated the classical pathway of complement but not the alternative pathway . At high antigen concentration the IgG2 antibody could also activate the classical pathway but the IgG4 and IgA2 antibodies could not . However the IgA2 antibody, and to a lesser extent the IgG2 antibody, did activate the alternative pathway of complement .