Symmetry and asymmetry in idiotypic interactions.

At the level of antibody molecules, the partners in an idiotypic interaction are functionally equivalent, and the distinction between "idiotype" and "anti-idiotype" is merely operational. This point must be taken into account in the interpretation of regulatory experiments with "idiotypic" and "antiidiotypic" antibodies. Indeed, consistent interpretations of certain sets of data can be obtained if "idiotype" and "antiidiotype" are used as operational terms for functionally equivalent molecules with complementary binding sites. For interacting antibody populations, symmetrical and asymmetrical situations can occur. The latter case is when an antibody family whose members express a set of idiotypes in various combinations--a typical situation in germ-line encoded immune responses--interacts with a heterogenous population of anti-idiotypes, the common denominators of which are binding sites for the common idiotopes expressed in the idiotypic antibody family. Through this type of interaction, one can explain in a first approximation the experiments of Urbain, Cazenave and others in which antiidiotypic antibody populations (ab2) induce antibodies (ab3) which resemble the original idiotypic population (ab1) idiotypically, but not primarily in antigen-binding specificity. Such a model would also describe an idiotypic network in which germline-encoded recurrent idiotypes (expressed in antibody families of the type described above) interact with anti-idiotypes generated by somatic mutation.