The concept of the idiotype of immunoglobulins emerged in the early 1960s from two different approaches, one by Oudin with rabbit antibodies, the other by Kunkel and analysis of the immunochemical characteristics of human myeloma proteins. At the time, Oudin had described the allotypic specificities as defining antigenic characteristics of a group of individuals within a given animal species. This was observed for rabbit immunoglobulins and was the consequence of allelic variants of both the heavy and light chains. It was then found that when rabbit antibodies prepared against Salmonella typhi were injected into another rabbit expressing the same (known) allotypes, they induced the synthesis of antibodies that specifically recognized the anti-Salmonella antibodies raised in the first rabbit. It was also shown that they did not react with normal rabbit serum taken before the Salmonella injection, thereby excluding that they identified a new allotypic specificity. Idiotypic specificities were thus defined as antigenic specificities characteristic of one antibody (the idiotype) produced by one animal and specific for one given antigen. Antibodies produced by the second animal were termed anti-idiotypic antibodies. Idiotypes are also referred to as Ab1 while anti-idiotypes are referred to as Ab2, in a simplified nomenclature.
At the same time, Kunkel was comparing the immunochemical characteristics of human myeloma proteins with those of normal immunoglobulins. He found that antibodies raised against one given myeloma protein, which cross-reacted against normal immunoglobulins, became specific for the immunizing protein once extensively adsorbed on normal immunoglobulins. This was taken as indicative of determinants specific for any given monoclonal product, although it was not entirely clear whether it was also linked to the pathological "abnormal" nature of the myeloma protein.
Structural analysis confirmed, as anticipated, that idiotypic determinants, or idiotopes, were present on the Fab fragments and, more precisely, on the variable regions of H and/or L chains (see Immunoglobulin Structure), depending on the idiotype. Extensive analysis of idiotype structure performed by a combination of immunochemical and structural analysis of anti-hapten antibodies revealed several types of idiotopes. Some Ab1-Ab2 interactions could be inhibited by a hapten of Ab1, indicating that the corresponding idiotopes were part of the antibody combining site, whereas other were not. Genetic analysis also revealed that some idiotypic specificities were common to several idiotypes, whereas others were strictly specific for one given Ab1, leading to the distinction of public and private idiotypes, a notion that was directly related to the dual origin of antibody diversity, germline encoded and somatically generated. Idiotypy was studied extensively when it was discovered that the cascade Ag(X) ^ Ab1 ^ Ab2 could be continued and amplified in a large idiotypic network of interactions, providing a basis for autoregulation of the immune system. An especially interesting observation was that Ab2 could induce certain Ab3 molecules that resembled Ab1 in that they could bind the original Ag(X) antigen. This led to the definition of the "internal image" of the antigen, proposed by Jerne, containing the idea that the collection of normal immunoglobulins of an individual could represent a huge population of internal images of the outside antigen world. It also stimulated approaches of idiotypic vaccines that could have been used in place of classical antigens, an interesting idea whenever antigens were either difficult to identify or poorly immunogenic.
