Evidence for multispecificity of antibody molecules

Abstract

ONE of the salient questions in immunology concerns the means by which the immune system is capable of recognising an apparently limitless variety of antigenic determinants, including many that presumably did not exist during its evolution. Three major hypotheses have been put forward to explain the diversity of the immune response: a stringent germ line hypothesis, a recombinational germ line hypothesis and a somatic mutation hypothesis (see ref. 1 for discussion). If the assumption is made that an antibody-combining site has a single antigenic specificity, the first of the above hypotheses postulates about 1,000 variable genes and 1,000 constant genes to code for about 10⁶ antigenic determinants. The second and, especially, the third hypotheses require the presence of many fewer genes in the germ line. Even fewer genes would be required if antibody-combining sites were multispecific. Evidence has been put forward to support the concept of multispecificity². Moreover, the ability of single myeloma globulins to bind structurally diverse molecules³ and the multispecific character of a Bence-Jones⁴ dimer can be taken as additional supportive evidence. In an earlier paper⁵, we attempted to investigate the question of multispecificity by studying the binding specificities of a population of antibodies of limited heterogeneity generated by immunisation with a conjugate of adenosine-5′-phosphate and gramicidin S: (AMP)₂–gramicidin S (ref. 6). Immunised rabbits produced predominantly three or four globulin proteins specific for AMP. Binding studies on the sera of one of the rabbits, in which there were three major AMP-specific globulins, yielded data that could be interpreted as evidence for multispecificity of antibody-combining sites. We have now succeeded in isolating three homogeneous anti-AMP antibodies from one of the rabbits immunised with (AMP)₂–gramicidin S and have studied the ability of each globulin to bind AMP and other ligands. The isoelectric pattern of the specifically purified⁷ population of antibodies obtained from the serum of rabbit 449 (ref. 5) showed three major bands. The three bands were separated and isolated by preparative isoelectric focusing in polyacrylamide by an adaption of the procedure of Wrigley⁸, and were shown to be homogeneous by analytical isoelectric focusing. They were designated fractions A, B and C.